https://blog.schoergendorfer.com/post/microsoft-named-a-leader-in-the-frost-radartm-cloud-workload-protection-platform-2SecFeedde2026-08-22T19:31:42.393289+00:00Microsoft named a Leader in the Frost Radar™: Cloud Workload Protection Platforms, 2026https://blog.schoergendorfer.com/post/llm-033SecFeedde2026-08-22T17:47:15.249954+00:00llm 0.33https://blog.schoergendorfer.com/post/the-open-mainframe-the-keystone-of-the-end-to-end-digital-enterpriseSecFeedde2026-08-22T17:01:45.954740+00:00The open mainframe: the keystone of the end-to-end digital enterprisehttps://blog.schoergendorfer.com/post/securing-sandboxes-what-happens-when-ai-agents-escape-containmentSecFeedde2026-08-22T16:16:35.266215+00:00Securing sandboxes: What happens when AI agents escape containment?https://blog.schoergendorfer.com/post/six-identity-capabilities-for-securing-autonomous-ai-agentsSecFeedde2026-08-22T14:16:25.432949+00:00Six identity capabilities for securing autonomous AI agentshttps://blog.schoergendorfer.com/post/decoding-ais-open-source-course-maps-three-ways-to-run-an-agent-loop-and-the-proSecFeedde2026-08-22T13:46:27.594535+00:00Decoding AI’s Open-Source Course Maps Three Ways to Run an Agent Loop and the Provider Economics Behind Eachhttps://blog.schoergendorfer.com/post/how-claude-watermarks-ai-generated-textSecFeedde2026-08-22T11:16:47.866826+00:00How Claude Watermarks AI-Generated Texthttps://blog.schoergendorfer.com/post/the-evolution-of-the-agent-harnessSecFeedde2026-08-22T07:46:25.710479+00:00The Evolution of the Agent Harnesshttps://blog.schoergendorfer.com/post/ainews-10-worse-100x-cheaper-10000x-faster-why-simulation-is-taking-overSecFeedde2026-08-22T07:46:25.682512+00:00[AINews] 10% worse, 100x cheaper, 10000x faster: Why Simulation is taking overhttps://blog.schoergendorfer.com/post/say-it-once-introducing-bot-preference-sync-3SecFeedde2026-08-22T02:31:35.252080+00:00Say it once: introducing Bot Preference Synchttps://blog.schoergendorfer.com/post/say-it-once-introducing-bot-preference-sync-2SecFeedde2026-08-21T23:46:25.976155+00:00Say it once: introducing Bot Preference SynChttps://blog.schoergendorfer.com/post/simulation-the-new-scaling-law-joon-sung-park-simile-aiSecFeedde2026-08-21T23:46:25.948723+00:00Simulation: the new Scaling Law — Joon Sung Park, Simile AIhttps://blog.schoergendorfer.com/post/say-it-once-introducing-bot-preference-syncSecFeedde2026-08-21T23:31:24.241177+00:00Say it once: introducing Bot Preference Synchttps://blog.schoergendorfer.com/post/best-gpu-neoclouds-2026-coreweave-nebius-lambda-crusoe-and-groq-ranked-by-publisSecFeedde2026-08-21T23:31:24.210430+00:00Best GPU Neoclouds 2026: CoreWeave, Nebius, Lambda, Crusoe, and Groq Ranked by Published Pricing and Contracted Powerhttps://blog.schoergendorfer.com/post/amazon-eks-capability-for-argo-cd-now-supports-custom-configurationSecFeedde2026-08-21T23:16:49.682560+00:00Amazon EKS Capability for Argo CD now supports custom configurationhttps://blog.schoergendorfer.com/post/building-agentic-document-intelligence-pipelines-creating-scientific-figures-witSecFeedde2026-08-21T22:16:48.341027+00:00Building Agentic Document Intelligence Pipelines: Creating Scientific Figures with AutoFigurehttps://blog.schoergendorfer.com/post/anthropic-brings-claude-mythos-5-to-claude-security-enterprise-teams-get-frontieSecFeedde2026-08-21T21:32:00.112497+00:00Anthropic Brings Claude Mythos 5 to Claude Security: Enterprise Teams Get Frontier Vulnerability Scanning Without Direct Model Accesshttps://blog.schoergendorfer.com/post/amazon-connect-customer-now-lets-managers-chat-with-their-dataSecFeedde2026-08-21T21:16:27.531474+00:00Amazon Connect Customer now lets managers chat with their datahttps://blog.schoergendorfer.com/post/friday-squid-blogging-neon-flying-squidSecFeedde2026-08-21T21:16:27.467134+00:00Friday Squid Blogging: Neon Flying Squidhttps://blog.schoergendorfer.com/post/spline-rebuilt-its-entire-3d-editor-then-it-handed-the-keys-to-claude-codeSecFeedde2026-08-21T20:31:39.071864+00:00Spline rebuilt its entire 3D editor. Then it handed the keys to Claude Code.https://blog.schoergendorfer.com/post/usn-8668-1-linux-kernel-gcp-vulnerabilitiesSecFeedde2026-08-21T20:16:42.220130+00:00USN-8668-1: Linux kernel (GCP) vulnerabilitieshttps://blog.schoergendorfer.com/post/usn-8669-1-linux-kernel-nvidia-vulnerabilitiesSecFeedde2026-08-21T20:16:42.137686+00:00USN-8669-1: Linux kernel (NVIDIA) vulnerabilitieshttps://blog.schoergendorfer.com/post/dash0-acquires-polar-signals-for-continuous-profiling-and-gpu-visibilitySecFeedde2026-08-21T20:16:41.341016+00:00Dash0 Acquires Polar Signals for Continuous Profiling and GPU Visibilityhttps://blog.schoergendorfer.com/post/transforming-mainframe-recoverySecFeedde2026-08-21T20:16:41.311572+00:00Transforming Mainframe Recoveryhttps://blog.schoergendorfer.com/post/91-spring-cves-the-ai-vulnerability-consumption-problemSecFeedde2026-08-21T20:02:07.285698+00:0091 Spring CVEs: The AI Vulnerability Consumption Problemhttps://blog.schoergendorfer.com/post/usn-8667-1-linux-kernel-kvm-vulnerabilitiesSecFeedde2026-08-21T19:46:34.500875+00:00USN-8667-1: Linux kernel (KVM) vulnerabilitieshttps://blog.schoergendorfer.com/post/anthropic-brings-mythos-5-to-its-claude-security-vulnerability-scannerSecFeedde2026-08-21T19:16:31.544356+00:00Anthropic brings Mythos 5 to its Claude Security vulnerability scannerhttps://blog.schoergendorfer.com/post/amazon-ses-now-supports-open-and-click-tracking-override-parametersSecFeedde2026-08-21T19:16:31.521162+00:00Amazon SES now supports open and click tracking override parametershttps://blog.schoergendorfer.com/post/aws-glue-60-delivers-30-price-reduction-and-iceberg-v3-supportSecFeedde2026-08-21T19:16:31.499776+00:00AWS Glue 6.0 delivers 30% price reduction and Iceberg v3 supporthttps://blog.schoergendorfer.com/post/aws-deadline-cloud-now-tracks-automatic-download-status-in-the-deadline-cloud-moSecFeedde2026-08-21T19:16:31.477034+00:00AWS Deadline Cloud now tracks automatic download status in the Deadline Cloud Monitorhttps://blog.schoergendorfer.com/post/aws-glue-60-now-available-with-30-lower-price-and-full-apache-iceberg-v3-supportSecFeedde2026-08-21T19:01:48.182242+00:00AWS Glue 6.0 now available with 30% lower price and full Apache Iceberg v3 supporthttps://blog.schoergendorfer.com/post/hackers-target-popular-arrayref-rust-crate-in-supply-chain-attackSecFeedde2026-08-21T18:46:42.568273+00:00Hackers Target Popular arrayref Rust Crate in Supply-Chain Attackhttps://blog.schoergendorfer.com/post/cisa-adds-one-known-exploited-vulnerability-to-catalog-11SecFeedde2026-08-21T18:46:42.538439+00:00CISA Adds One Known Exploited Vulnerability to Cataloghttps://blog.schoergendorfer.com/post/build-a-unified-ai-agent-architecture-with-dynamodb-and-bedrockSecFeedde2026-08-21T18:31:44.574995+00:00Build a unified AI agent architecture with DynamoDB and Bedrockhttps://blog.schoergendorfer.com/post/amazon-ec2-c8gd-m8gd-and-r8gd-instances-are-now-available-in-additional-aws-regiSecFeedde2026-08-21T18:16:29.460495+00:00Amazon EC2 C8gd, M8gd and R8gd instances are now available in additional AWS Regionshttps://blog.schoergendorfer.com/post/govern-ai-agent-tool-access-with-amazon-bedrock-agentcore-gatewaySecFeedde2026-08-21T17:16:53.868900+00:00Govern AI agent tool access with Amazon Bedrock AgentCore Gatewayhttps://blog.schoergendorfer.com/post/agentic-data-operations-platform-adop-data-engineering-into-hoursSecFeedde2026-08-21T17:16:53.831572+00:00Agentic Data Operations Platform (ADOP): Data engineering into hourshttps://blog.schoergendorfer.com/post/anthropics-new-browser-tool-doesnt-actually-run-a-browserSecFeedde2026-08-21T17:01:32.733381+00:00Anthropic’s new browser tool doesn’t actually run a browserhttps://blog.schoergendorfer.com/post/accelerating-aircraft-ifec-diagnostics-with-agentic-ai-on-awsSecFeedde2026-08-21T17:01:32.706026+00:00Accelerating aircraft IFEC diagnostics with agentic AI on AWShttps://blog.schoergendorfer.com/post/reduce-rag-costs-on-amazon-bedrock-with-query-aware-compressionSecFeedde2026-08-21T17:01:32.678208+00:00Reduce RAG costs on Amazon Bedrock with query-aware compressionhttps://blog.schoergendorfer.com/post/ai-is-learning-to-write-genetic-codeSecFeedde2026-08-21T17:01:31.887699+00:00AI Is Learning to Write Genetic Codehttps://blog.schoergendorfer.com/post/how-agents-can-delegate-betterSecFeedde2026-08-21T16:16:43.577043+00:00How agents can delegate betterhttps://blog.schoergendorfer.com/post/cloud-ciso-perspectives-sticking-to-security-fundamentals-in-the-ai-eraSecFeedde2026-08-21T16:16:43.547261+00:00Cloud CISO Perspectives: Sticking to security fundamentals in the AI erahttps://blog.schoergendorfer.com/post/forget-the-model-wars-stripe-and-ramp-just-started-the-router-warsSecFeedde2026-08-21T14:31:34.824850+00:00Forget the model wars, Stripe and Ramp just started the router warshttps://blog.schoergendorfer.com/post/claude-opus-5-scored-30-on-arc-agi-3-wrapped-in-nvidias-avo-it-hit-100SecFeedde2026-08-21T13:16:16.887616+00:00Claude Opus 5 scored 30% on ARC-AGI-3. Wrapped in Nvidia’s AVO, it hit 100%.https://blog.schoergendorfer.com/post/grok-claude-and-hermes-agents-get-job-titles-and-persistent-permissionsSecFeedde2026-08-21T13:16:16.861093+00:00Grok, Claude, and Hermes agents get job titles — and persistent permissionshttps://blog.schoergendorfer.com/post/running-ai-agents-in-github-actions-with-docker-sandboxesSecFeedde2026-08-21T13:01:43.011347+00:00Running AI agents in GitHub Actions with Docker Sandboxeshttps://blog.schoergendorfer.com/post/production-grade-ai-eval-systems-what-i-learned-putting-llms-on-callSecFeedde2026-08-21T12:31:42.793297+00:00Production-Grade AI Eval Systems. What I Learned Putting LLMs on Callhttps://blog.schoergendorfer.com/post/production-grade-ai-eval-systemsSecFeedde2026-08-21T12:16:47.567331+00:00Production-Grade AI Eval Systemshttps://blog.schoergendorfer.com/post/most-coding-agent-benchmarks-skip-large-scale-refactoring-not-this-oneSecFeedde2026-08-21T12:01:31.521327+00:00Most coding agent benchmarks skip large-scale refactoring. Not this one.https://blog.schoergendorfer.com/post/waterfall-20-controlling-llm-driven-software-development-with-stage-gated-discipSecFeedde2026-08-21T11:46:29.372461+00:00Waterfall 2.0: Controlling LLM-Driven Software Development with Stage-Gated Disciplinehttps://blog.schoergendorfer.com/post/how-to-turn-slow-queries-into-actionable-reliability-metrics-with-opentelemetrySecFeedde2026-08-21T11:16:26.599214+00:00How to turn slow queries into actionable reliability metrics with OpenTelemetryhttps://blog.schoergendorfer.com/post/from-clickops-to-governed-iac-cloudformation-drift-detection-in-practiceSecFeedde2026-08-21T11:01:34.245748+00:00From clickops to governed IaC: CloudFormation drift detection in practicehttps://blog.schoergendorfer.com/post/why-api-test-generation-is-a-judgment-problem-not-a-code-generation-problemSecFeedde2026-08-21T10:46:30.617116+00:00Why API Test Generation Is a Judgment Problem, Not a Code Generation Problemhttps://blog.schoergendorfer.com/post/why-cryptographic-inventory-is-the-first-step-toward-quantum-readinessSecFeedde2026-08-21T10:31:55.524346+00:00Why Cryptographic Inventory Is the First Step Toward Quantum Readinesshttps://blog.schoergendorfer.com/post/how-agentflo-built-ai-sales-agents-with-amazon-bedrock-agentcore-part-2SecFeedde2026-08-21T10:31:55.477756+00:00How AgentFlo built AI sales agents with Amazon Bedrock AgentCore – Part 2https://blog.schoergendorfer.com/post/more-incidents-of-ais-going-rogue-in-cybersecurity-challengesSecFeedde2026-08-21T10:16:21.117381+00:00More Incidents of AIs Going Rogue in Cybersecurity Challengeshttps://blog.schoergendorfer.com/post/the-invisible-passenger-in-your-carSecFeedde2026-08-21T09:46:28.282751+00:00The invisible passenger in your carhttps://blog.schoergendorfer.com/post/ainews-poolside-gets-12b-reverse-execuhire-to-nvidia-founders-stay-for-1b-employSecFeedde2026-08-21T09:16:25.157999+00:00[AINews] Poolside gets $12B reverse-execuhire to NVIDIA; founders stay for $1B, employees go for $6B, Infraco scaling to 7GW neocloudhttps://blog.schoergendorfer.com/post/greekbarretrieval-a-benchmark-for-greek-statutory-retrieval-2SecFeedde2026-08-21T09:01:38.621989+00:00GreekBarRetrieval: A Benchmark for Greek Statutory Retrievalhttps://blog.schoergendorfer.com/post/contesttrade-a-multi-agent-trading-system-based-on-internal-contest-mechanism-2SecFeedde2026-08-21T09:01:38.584759+00:00ContestTrade: A Multi-Agent Trading System Based on Internal Contest Mechanismhttps://blog.schoergendorfer.com/post/mitigating-genai-powered-evidence-pollution-for-out-of-context-misinformation-deSecFeedde2026-08-21T09:01:38.563429+00:00Mitigating GenAI-Powered Evidence Pollution for Out-Of-Context Misinformation Detectionhttps://blog.schoergendorfer.com/post/grading-the-graders-verification-autonomy-levels-l0-l5-for-llm-reasoning-2SecFeedde2026-08-21T09:01:38.545182+00:00Grading the Graders: Verification Autonomy Levels (L0-L5) for LLM Reasoninghttps://blog.schoergendorfer.com/post/spyce-skill-policy-co-evolution-for-multimodal-agentsSecFeedde2026-08-21T09:01:38.508311+00:00SPyCE: Skill-Policy Co-evolution for Multimodal Agentshttps://blog.schoergendorfer.com/post/repselect-robust-llm-unlearning-via-representation-selectivitySecFeedde2026-08-21T09:01:38.487076+00:00RepSelect: Robust LLM Unlearning via Representation Selectivityhttps://blog.schoergendorfer.com/post/calibrated-triage-not-autonomy-confidence-estimation-for-medical-vision-languageSecFeedde2026-08-21T09:01:38.459196+00:00Calibrated Triage, Not Autonomy: Confidence Estimation for Medical Vision-Language Modelshttps://blog.schoergendorfer.com/post/self-harness-harnesses-that-improve-themselves-2SecFeedde2026-08-21T08:46:33.265828+00:00Self-Harness: Harnesses That Improve Themselveshttps://blog.schoergendorfer.com/post/a-finite-calibration-regime-map-for-llm-judge-panelsSecFeedde2026-08-21T08:46:33.245173+00:00A Finite-Calibration Regime Map for LLM Judge Panelshttps://blog.schoergendorfer.com/post/evoselect-data-efficient-llm-evolution-for-targeted-task-adaptationSecFeedde2026-08-21T08:46:33.226042+00:00EvoSelect: Data-Efficient LLM Evolution for Targeted Task Adaptationhttps://blog.schoergendorfer.com/post/remask-dont-replace-token-to-mask-refinement-in-diffusion-language-modelsSecFeedde2026-08-21T08:46:33.200764+00:00Remask, Don't Replace: Token-to-Mask Refinement in Diffusion Language Modelshttps://blog.schoergendorfer.com/post/when-contextual-inference-fails-cancelability-in-interactive-instruction-followiSecFeedde2026-08-21T08:46:33.164280+00:00When Contextual Inference Fails: Cancelability in Interactive Instruction Followinghttps://blog.schoergendorfer.com/post/multi-agent-orchestration-with-the-common-sense-reasoning-capabilities-of-llms-fSecFeedde2026-08-21T08:46:33.127528+00:00Multi-Agent Orchestration with the Common-Sense Reasoning Capabilities of LLMs for Autonomous Drivinghttps://blog.schoergendorfer.com/post/towards-clinically-faithful-medical-image-captioning-via-enhanced-vision-languagSecFeedde2026-08-21T08:46:33.105692+00:00Towards Clinically Faithful Medical Image Captioning via Enhanced Vision-Language Alignmenthttps://blog.schoergendorfer.com/post/streamsoccer-event-driven-memory-for-streaming-soccer-commentarySecFeedde2026-08-21T08:46:33.083220+00:00StreamSoccer: Event-Driven Memory for Streaming Soccer Commentaryhttps://blog.schoergendorfer.com/post/harp-hierarchical-adaptive-ranking-with-preference-adaptive-fusion-for-query-basSecFeedde2026-08-21T08:46:33.043095+00:00HARP: Hierarchical Adaptive Ranking with Preference-Adaptive Fusion for Query-Based CVE Prioritizationhttps://blog.schoergendorfer.com/post/conceptguard-benchmarking-context-sensitive-unlearning-in-large-language-modelsSecFeedde2026-08-21T08:46:33.022003+00:00ConceptGuard: Benchmarking Context-Sensitive Unlearning in Large Language Modelshttps://blog.schoergendorfer.com/post/formaltcs-benchmarking-end-to-end-frontier-formal-theoretical-computer-science-rSecFeedde2026-08-21T08:46:32.999327+00:00FormalTCS: Benchmarking End-to-End Frontier Formal Theoretical Computer Science Research of Large Language Modelshttps://blog.schoergendorfer.com/post/when-text-and-numbers-disagree-evidence-arbitration-in-large-language-modelsSecFeedde2026-08-21T08:46:32.980004+00:00When Text and Numbers Disagree: Evidence Arbitration in Large Language Modelshttps://blog.schoergendorfer.com/post/oenobench-a-wine-domain-benchmark-for-knowledge-grounded-evaluation-of-large-lanSecFeedde2026-08-21T08:46:32.958063+00:00OenoBench: A Wine-Domain Benchmark for Knowledge-Grounded Evaluation of Large Language Modelshttps://blog.schoergendorfer.com/post/robust-incomplete-multimodal-sentiment-analysis-via-iterative-proxy-correctionSecFeedde2026-08-21T08:46:32.920729+00:00Robust Incomplete Multimodal Sentiment Analysis via Iterative Proxy Correctionhttps://blog.schoergendorfer.com/post/natural-language-code-retrieval-for-1centerprise-an-open-benchmark-and-efficientSecFeedde2026-08-21T08:46:32.900620+00:00Natural Language Code Retrieval for 1C:Enterprise: An Open Benchmark and Efficient Bi-Encoderhttps://blog.schoergendorfer.com/post/dynamic-gated-cross-modal-fusion-with-sarcastic-aware-contrastive-regularizationSecFeedde2026-08-21T08:46:32.875718+00:00Dynamic Gated Cross-Modal Fusion with Sarcastic-aware Contrastive Regularization for Multimodal Sarcasm Detectionhttps://blog.schoergendorfer.com/post/stopping-and-routing-llm-judge-panelsSecFeedde2026-08-21T08:31:33.368619+00:00Stopping and Routing LLM Judge Panelshttps://blog.schoergendorfer.com/post/swe-bench-science-can-coding-agents-resolve-engineering-tasks-in-scienceSecFeedde2026-08-21T08:31:33.343770+00:00SWE-bench Science: Can Coding Agents Resolve Engineering Tasks in Science?https://blog.schoergendorfer.com/post/flashprefill-v2-block-sparse-prefill-attention-for-long-context-llm-servingSecFeedde2026-08-21T08:31:33.322878+00:00FlashPrefill V2: Block-Sparse Prefill Attention for Long-Context LLM Servinghttps://blog.schoergendorfer.com/post/personalbench-measuring-the-authorship-gap-in-llm-personalizationSecFeedde2026-08-21T08:31:33.301034+00:00PersonalBench: Measuring the Authorship Gap in LLM Personalizationhttps://blog.schoergendorfer.com/post/one-success-isnt-reliability-thinkingbox-a-sandbox-and-benchmark-for-agents-in-sSecFeedde2026-08-21T08:31:33.281086+00:00One Success Isn't Reliability: Thinkingbox, a Sandbox and Benchmark for Agents in Stateful Business Workflowshttps://blog.schoergendorfer.com/post/recache-efficient-kv-cache-reuse-and-compression-for-tool-augmented-llm-agentsSecFeedde2026-08-21T08:31:33.244116+00:00ReCache: Efficient KV Cache Reuse and Compression for Tool-Augmented LLM Agentshttps://blog.schoergendorfer.com/post/mitigating-identity-essentialism-in-llm-agents-with-longitudinal-life-trajectoriSecFeedde2026-08-21T08:31:33.226163+00:00Mitigating Identity Essentialism in LLM Agents with Longitudinal Life Trajectorieshttps://blog.schoergendorfer.com/post/remember-verify-or-ask-cross-family-evaluation-of-memory-commitment-in-llm-agentSecFeedde2026-08-21T08:31:33.204283+00:00Remember, Verify, or Ask? Cross-Family Evaluation of Memory Commitment in LLM Agentshttps://blog.schoergendorfer.com/post/reliable-financial-named-entity-recognition-under-domain-shiftSecFeedde2026-08-21T08:31:33.184014+00:00Reliable Financial Named Entity Recognition under Domain Shifthttps://blog.schoergendorfer.com/post/generating-diverse-personas-for-user-simulators-to-test-interview-dialogue-systeSecFeedde2026-08-21T08:31:33.166324+00:00Generating Diverse Personas for User Simulators to Test Interview Dialogue Systemshttps://blog.schoergendorfer.com/post/hear2act-benchmarking-when-prosody-should-change-what-an-assistant-doesSecFeedde2026-08-21T08:31:33.147286+00:00Hear2Act: Benchmarking When Prosody Should Change What an Assistant Doeshttps://blog.schoergendorfer.com/post/synflow-a-multidimensional-diachronic-semantic-analysis-toolkitSecFeedde2026-08-21T08:31:33.128650+00:00SynFlow: A Multidimensional Diachronic Semantic Analysis Toolkithttps://blog.schoergendorfer.com/post/linguistic-holonomy-and-statistical-watermarks-inner-geometry-of-meaning-preservSecFeedde2026-08-21T08:31:33.107705+00:00Linguistic Holonomy and Statistical Watermarks: Inner Geometry of Meaning-Preserving Transformationshttps://blog.schoergendorfer.com/post/nepooc-m-bilingual-nepali-english-benchmark-and-comparative-analysis-of-multimodSecFeedde2026-08-21T08:31:33.063704+00:00NepOOC-M: Bilingual Nepali-English Benchmark and Comparative Analysis of Multimodal Architectures for OOC Detectionhttps://blog.schoergendorfer.com/post/when-irrelevant-text-matters-affine-margin-shifts-in-multimodal-large-language-mSecFeedde2026-08-21T08:31:33.044229+00:00When Irrelevant Text Matters: Affine Margin Shifts in Multimodal Large Language Modelshttps://blog.schoergendorfer.com/post/compliance-capability-and-conflict-benchmarking-multimodal-llms-under-system-mesSecFeedde2026-08-21T08:31:33.022915+00:00Compliance, Capability, and Conflict: Benchmarking Multimodal LLMs under System Messageshttps://blog.schoergendorfer.com/post/compress-and-forget-bitsandbytes-quantization-amplifies-proactive-interference-i-2SecFeedde2026-08-21T08:31:32.998599+00:00Compress and Forget: bitsandbytes Quantization Amplifies Proactive Interference in LLMshttps://blog.schoergendorfer.com/post/automating-learner-assessment-benchmarking-machine-learning-and-deep-learning-mo-2SecFeedde2026-08-21T08:17:10.509637+00:00Automating Learner Assessment: Benchmarking Machine Learning and Deep Learning Models for EEG-Based Familiarity Predictionhttps://blog.schoergendorfer.com/post/lodestar-robust-entropy-based-answer-selection-in-retrieval-augmented-generationSecFeedde2026-08-21T08:17:10.488324+00:00LODESTAR: Robust Entropy-Based Answer Selection in Retrieval-Augmented Generation for Question Answering -- Directing Frozen-LLM Entropy with a Reinforcement-Learned Prompt Polarizer under Misleading Passageshttps://blog.schoergendorfer.com/post/uncertainty-aware-compositional-localization-and-placement-assessment-of-cathete-2SecFeedde2026-08-21T08:17:10.447731+00:00Uncertainty-Aware Compositional Localization and Placement Assessment of Catheters and Tubes in Chest X-Rayshttps://blog.schoergendorfer.com/post/provably-efficient-self-calibrating-quantum-fault-tolerance-2SecFeedde2026-08-21T08:17:10.427507+00:00Provably Efficient Self-Calibrating Quantum Fault Tolerancehttps://blog.schoergendorfer.com/post/learning-asymptotics-with-convergence-rate-guarantees-using-linear-least-squares-2SecFeedde2026-08-21T08:17:10.408643+00:00Learning Asymptotics with Convergence-Rate Guarantees using Linear Least Squareshttps://blog.schoergendorfer.com/post/the-price-of-hidden-curvature-improved-lower-bounds-for-bandit-convex-optimizatiSecFeedde2026-08-21T08:17:10.388235+00:00The Price of Hidden Curvature: Improved Lower Bounds for Bandit Convex Optimizationhttps://blog.schoergendorfer.com/post/distributional-determinantal-point-process-for-repulsive-clustering-of-distribut-2SecFeedde2026-08-21T08:17:10.365147+00:00Distributional Determinantal Point Process for Repulsive Clustering of Distributionshttps://blog.schoergendorfer.com/post/continuous-behavioral-authentication-via-multi-expert-bert-log-analysis-for-secuSecFeedde2026-08-21T08:17:10.324898+00:00Continuous Behavioral Authentication via Multi-Expert BERT Log Analysis for Secure Data Sharinghttps://blog.schoergendorfer.com/post/the-value-of-finite-observation-in-positive-data-learning-of-multiple-context-frSecFeedde2026-08-21T08:17:10.286675+00:00The Value of Finite Observation in Positive-Data Learning of Multiple Context-Free Languageshttps://blog.schoergendorfer.com/post/learn-for-variation-efficient-aav-trajectory-learning-through-a-differentiable-wSecFeedde2026-08-21T08:17:10.265472+00:00Learn for Variation: Efficient AAV Trajectory Learning through a Differentiable Wireless World Modelhttps://blog.schoergendorfer.com/post/reducing-the-complexity-of-matrix-multiplication-by-quantum-computingSecFeedde2026-08-21T08:17:10.226200+00:00Reducing the Complexity of Matrix Multiplication by Quantum Computinghttps://blog.schoergendorfer.com/post/better-call-graphs-a-new-dataset-of-function-call-graphs-for-malware-classificatSecFeedde2026-08-21T08:17:10.202945+00:00Better Call Graphs: A New Dataset of Function Call Graphs for Malware Classificationhttps://blog.schoergendorfer.com/post/decoupling-high-and-low-frequencies-for-faithful-image-generation-with-fine-detaSecFeedde2026-08-21T08:17:10.179659+00:00Decoupling High and Low Frequencies for Faithful Image Generation with Fine Detailshttps://blog.schoergendorfer.com/post/asymptotic-theory-for-iv-based-reinforcement-learning-with-potential-endogeneitySecFeedde2026-08-21T08:17:10.155630+00:00Asymptotic Theory for IV-Based Reinforcement Learning with Potential Endogeneityhttps://blog.schoergendorfer.com/post/continual-reasoning-gym-diagnosing-and-harnessing-shared-reasoning-in-continual-2SecFeedde2026-08-21T08:17:10.133350+00:00Continual Reasoning Gym: Diagnosing and Harnessing Shared Reasoning in Continual RLVRhttps://blog.schoergendorfer.com/post/geometric-and-behavioral-stratification-in-transformer-residual-streams-3SecFeedde2026-08-21T08:17:10.106340+00:00Geometric and Behavioral Stratification in Transformer Residual Streamshttps://blog.schoergendorfer.com/post/agents-unlock-new-capabilities-through-switching-lora-adapters-as-a-tool-slaaat-2SecFeedde2026-08-21T08:01:58.027217+00:00Agents unlock new capabilities through Switching LoRA Adapters as a Tool (SLAaaT)https://blog.schoergendorfer.com/post/dash-opd-discrepancy-aware-switching-with-hysteresis-for-on-policy-distillation-2SecFeedde2026-08-21T08:01:57.956213+00:00DASH-OPD: Discrepancy-Aware Switching with Hysteresis for On-Policy Distillationhttps://blog.schoergendorfer.com/post/grounded-verification-of-chemical-and-materials-reasoning-detection-is-the-bottl-2SecFeedde2026-08-21T08:01:57.937444+00:00Grounded verification of chemical and materials reasoning: detection is the bottleneckhttps://blog.schoergendorfer.com/post/generalist-vision-language-models-for-fast-radio-burst-detection-a-zero-shot-ben-2SecFeedde2026-08-21T08:01:57.918286+00:00Generalist Vision-Language Models for Fast Radio Burst detection: a zero-shot benchmark against a specialized detectorhttps://blog.schoergendorfer.com/post/detecting-explanatory-insufficiency-in-learned-representations-a-framework-for-r-3SecFeedde2026-08-21T08:01:57.899734+00:00Detecting Explanatory Insufficiency in Learned Representations: A Framework for Representational Vigilancehttps://blog.schoergendorfer.com/post/bootstrap-theory-of-representational-emergence-tber-explanatory-insufficiency-trSecFeedde2026-08-21T08:01:57.861736+00:00Bootstrap Theory of Representational Emergence (TBER): Explanatory Insufficiency, Transition Regimes, and the Emergence of New Representational Levelshttps://blog.schoergendorfer.com/post/from-prediction-to-self-developmental-conditions-for-agency-in-minimal-neural-sySecFeedde2026-08-21T08:01:57.837376+00:00From Prediction to Self: Developmental Conditions for Agency in Minimal Neural Systemshttps://blog.schoergendorfer.com/post/requestrouter-request-boundary-routing-for-efficient-single-gpu-llm-inferenceSecFeedde2026-08-21T08:01:57.811170+00:00RequestRouter: Request-Boundary Routing for Efficient Single-GPU LLM Inferencehttps://blog.schoergendorfer.com/post/higher-resolution-better-generalization-unlocking-visual-scaling-in-deep-reinforSecFeedde2026-08-21T08:01:57.790627+00:00Higher Resolution, Better Generalization: Unlocking Visual Scaling in Deep Reinforcement Learninghttps://blog.schoergendorfer.com/post/transformer-see-transformer-do-copying-as-an-intermediate-step-in-learning-analoSecFeedde2026-08-21T08:01:57.748643+00:00Transformer See, Transformer Do: Copying as an Intermediate Step in Learning Analogical Reasoninghttps://blog.schoergendorfer.com/post/from-rebound-to-remedy-understanding-and-mitigating-reward-hacking-via-represent-2SecFeedde2026-08-21T08:01:57.726854+00:00From Rebound to Remedy: Understanding and Mitigating Reward Hacking via Representation Engineeringhttps://blog.schoergendorfer.com/post/neural-prior-estimation-learning-class-priors-from-latent-representationsSecFeedde2026-08-21T08:01:57.705505+00:00Neural Prior Estimation: Learning Class Priors from Latent Representationshttps://blog.schoergendorfer.com/post/guided-diffusion-by-optimized-loss-functions-on-relaxed-parameters-for-inverse-mSecFeedde2026-08-21T08:01:57.683833+00:00Guided Diffusion by Optimized Loss Functions on Relaxed Parameters for Inverse Material Designhttps://blog.schoergendorfer.com/post/maximum-likelihood-reinforcement-learningSecFeedde2026-08-21T08:01:57.663581+00:00Maximum Likelihood Reinforcement Learninghttps://blog.schoergendorfer.com/post/hbvla-pushing-1-bit-post-training-quantization-for-vision-language-action-modelsSecFeedde2026-08-21T08:01:57.642448+00:00HBVLA: Pushing 1-Bit Post-Training Quantization for Vision-Language-Action Modelshttps://blog.schoergendorfer.com/post/partition-of-unity-neural-networks-for-interpretable-classification-with-expliciSecFeedde2026-08-21T07:46:28.232797+00:00Partition of Unity Neural Networks for Interpretable Classification with Explicit Class Regionshttps://blog.schoergendorfer.com/post/carbench-a-comprehensive-benchmark-for-neural-surrogates-on-high-fidelity-3d-carSecFeedde2026-08-21T07:46:28.210312+00:00CarBench: A Comprehensive Benchmark for Neural Surrogates on High-Fidelity 3D Car Aerodynamicshttps://blog.schoergendorfer.com/post/renyi-sharpness-a-novel-sharpness-that-strongly-correlates-with-generalizationSecFeedde2026-08-21T07:46:28.185390+00:00R\'enyi Sharpness: A Novel Sharpness that Strongly Correlates with Generalizationhttps://blog.schoergendorfer.com/post/merge-now-regret-later-the-hidden-cost-of-model-merging-is-adversarial-transferaSecFeedde2026-08-21T07:46:28.143665+00:00Merge Now, Regret Later: The Hidden Cost of Model Merging Is Adversarial Transferabilityhttps://blog.schoergendorfer.com/post/graphpfn-a-prior-data-fitted-graph-foundation-modelSecFeedde2026-08-21T07:46:28.121147+00:00GraphPFN: A Prior-Data Fitted Graph Foundation Modelhttps://blog.schoergendorfer.com/post/why-cant-i-see-my-clusters-a-precision-recall-approach-to-dimensionality-reductiSecFeedde2026-08-21T07:46:28.097592+00:00Why Can't I See My Clusters? A Precision-Recall Approach to Dimensionality Reduction Validationhttps://blog.schoergendorfer.com/post/proteinzero-self-improving-protein-generation-via-online-reinforcement-learningSecFeedde2026-08-21T07:46:28.075172+00:00ProteinZero: Self-Improving Protein Generation via Online Reinforcement Learninghttps://blog.schoergendorfer.com/post/deepconvcontext-a-multi-scale-approach-to-timeseries-classification-in-human-act-2SecFeedde2026-08-21T07:46:28.051012+00:00DeepConvContext: A Multi-Scale Approach to Timeseries Classification in Human Activity Recognitionhttps://blog.schoergendorfer.com/post/reaugment-model-zoo-guided-rl-for-few-shot-time-series-augmentation-and-forecastSecFeedde2026-08-21T07:46:28.011098+00:00ReAugment: Model Zoo-Guided RL for Few-Shot Time Series Augmentation and Forecastinghttps://blog.schoergendorfer.com/post/multi-modal-graph-interaction-for-multi-graph-convolution-network-in-urban-spatiSecFeedde2026-08-21T07:46:27.985951+00:00Multi-Modal Graph Interaction for Multi-Graph Convolution Network in Urban Spatiotemporal Forecastinghttps://blog.schoergendorfer.com/post/information-on-trajectories-martingales-and-random-timesSecFeedde2026-08-21T07:46:27.945287+00:00Information on trajectories: martingales and random timeshttps://blog.schoergendorfer.com/post/tcp_alpha-margin-controlled-confidence-estimation-for-reliable-music-informationSecFeedde2026-08-21T07:46:27.923807+00:00$TCP_\alpha$: Margin-Controlled Confidence estimation for reliable Music Information Retrievalhttps://blog.schoergendorfer.com/post/which-eviction-policy-should-an-llm-cache-use-a-systematic-study-across-workloadSecFeedde2026-08-21T07:46:27.901093+00:00Which Eviction Policy Should an LLM Cache Use? A Systematic Study Across Workloads, Capacities, and Encodershttps://blog.schoergendorfer.com/post/transfer-learning-in-nonparametric-regression-with-deep-relu-networksSecFeedde2026-08-21T07:46:27.879183+00:00Transfer Learning in Nonparametric Regression with Deep ReLU Networkshttps://blog.schoergendorfer.com/post/gravitational-wave-parameter-estimation-with-machine-learning-generated-surrogatSecFeedde2026-08-21T07:46:27.856154+00:00Gravitational-wave parameter estimation with machine-learning generated surrogate waveformshttps://blog.schoergendorfer.com/post/feature-evolution-and-migration-during-vision-transformer-trainingSecFeedde2026-08-21T07:46:27.833223+00:00Feature Evolution and Migration during Vision Transformer Traininghttps://blog.schoergendorfer.com/post/auditing-cross-lingual-fairness-in-language-model-watermarkingSecFeedde2026-08-21T07:31:28.394842+00:00Auditing Cross-Lingual Fairness in Language Model Watermarkinghttps://blog.schoergendorfer.com/post/from-street-view-imagery-to-street-quality-indicators-vision-language-inferenceSecFeedde2026-08-21T07:31:28.369681+00:00From Street View Imagery to Street Quality Indicators: Vision Language Inference for the Suburban 15-minute Cityhttps://blog.schoergendorfer.com/post/flow-matching-meets-3d-curvilinear-structure-segmentation-in-medical-imagingSecFeedde2026-08-21T07:31:28.344639+00:00Flow Matching Meets 3D Curvilinear Structure Segmentation in Medical Imaginghttps://blog.schoergendorfer.com/post/from-noise-to-signal-improving-security-log-anomaly-detection-using-llms-with-enSecFeedde2026-08-21T07:31:28.320113+00:00From Noise to Signal: Improving Security Log Anomaly Detection Using LLMs with Endpoint-Specific Logshttps://blog.schoergendorfer.com/post/a-layered-simplex-architecture-for-large-alphabetsSecFeedde2026-08-21T07:31:28.296804+00:00A Layered Simplex Architecture for Large Alphabetshttps://blog.schoergendorfer.com/post/the-impact-of-feature-engineering-and-an-optimisation-framework-for-ocean-colourSecFeedde2026-08-21T07:31:28.248765+00:00The impact of feature engineering and an optimisation framework for ocean colour machine learninghttps://blog.schoergendorfer.com/post/a-repeated-measurements-approach-to-soh-battery-modelling-of-cyclic-aged-data-inSecFeedde2026-08-21T07:31:28.199893+00:00A Repeated Measurements Approach to $SoH$ Battery Modelling of Cyclic Aged Data in a Laboratory Environmenthttps://blog.schoergendorfer.com/post/learning-piecewise-smooth-dynamical-systemsSecFeedde2026-08-21T07:31:28.173014+00:00Learning piecewise-smooth dynamical systemshttps://blog.schoergendorfer.com/post/far-from-the-crowd-scalable-self-supervised-learning-via-geographic-isolationSecFeedde2026-08-21T07:31:28.128215+00:00Far from the Crowd: Scalable Self-Supervised Learning via Geographic Isolationhttps://blog.schoergendorfer.com/post/ripe-reinforced-keypoint-learning-from-positive-pairs-onlySecFeedde2026-08-21T07:31:28.085417+00:00RIPE++: Reinforced Keypoint Learning from Positive Pairs Onlyhttps://blog.schoergendorfer.com/post/learning-deterministic-and-stochastic-forced-hamiltonian-systemsSecFeedde2026-08-21T07:31:28.062202+00:00Learning Deterministic and Stochastic Forced Hamiltonian Systemshttps://blog.schoergendorfer.com/post/cacheroute-planned-prefix-affinity-routing-for-large-scale-llm-servingSecFeedde2026-08-21T07:31:28.039408+00:00CacheRoute: Planned Prefix-Affinity Routing for Large-Scale LLM Servinghttps://blog.schoergendorfer.com/post/scaffolding-minds-optimizing-latent-visual-target-representations-for-multimodalSecFeedde2026-08-21T07:31:28.018165+00:00Scaffolding Minds: Optimizing Latent Visual Target Representations for Multimodal Reasoninghttps://blog.schoergendorfer.com/post/composition-driven-phase-evolution-in-sm-doped-bifeo3-via-latent-field-reconstruSecFeedde2026-08-21T07:31:27.993109+00:00Composition-Driven Phase Evolution in Sm-Doped BiFeO3 via Latent-Field Reconstruction of Atomically Resolved STEM Datahttps://blog.schoergendorfer.com/post/fine-tuning-vlas-with-self-demonstrated-generative-control-for-multi-task-manipuSecFeedde2026-08-21T07:31:27.967400+00:00Fine-Tuning VLAs with Self-Demonstrated Generative Control for Multi-Task Manipulationhttps://blog.schoergendorfer.com/post/microlensify-a-transformer-based-machine-learning-classifier-for-microlensing-evSecFeedde2026-08-21T07:16:24.505802+00:00Microlensify: a Transformer Based Machine Learning Classifier for Microlensing Events Trained on TESS Light Curveshttps://blog.schoergendorfer.com/post/deep-mkv-ts-path-dependent-mckean-vlasov-control-for-financial-time-series-generSecFeedde2026-08-21T07:16:24.484070+00:00Deep-MKV-TS: Path-Dependent McKean--Vlasov Control for Financial Time Series Generationhttps://blog.schoergendorfer.com/post/data-driven-time-varying-control-barrier-functions-for-adaptive-safe-set-learninSecFeedde2026-08-21T07:16:24.443782+00:00Data-Driven Time-Varying Control Barrier Functions for Adaptive Safe-Set Learning with Online Decremental Support Vector Machineshttps://blog.schoergendorfer.com/post/deep-neural-networks-as-lattice-gauge-theoriesSecFeedde2026-08-21T07:16:24.424573+00:00Deep neural networks as lattice gauge theorieshttps://blog.schoergendorfer.com/post/quantum-gaussian-processes-for-prediction-of-channel-observationsSecFeedde2026-08-21T07:16:24.403552+00:00Quantum Gaussian processes for prediction of channel observationshttps://blog.schoergendorfer.com/post/clustering-and-token-denoising-for-faster-and-more-robust-vlmsSecFeedde2026-08-21T07:16:24.381758+00:00Clustering and Token Denoising for Faster and More Robust VLMshttps://blog.schoergendorfer.com/post/recovering-nonlinear-functions-of-latent-variables-a-plausible-value-neural-netwSecFeedde2026-08-21T07:16:24.362664+00:00Recovering Nonlinear Functions of Latent Variables: A Plausible-Value Neural Network Frameworkhttps://blog.schoergendorfer.com/post/demons-on-a-budget-adaptive-measurement-placement-at-the-entanglement-phase-tranSecFeedde2026-08-21T07:16:24.341344+00:00Demons on a Budget: Adaptive Measurement Placement at the Entanglement Phase Transitionhttps://blog.schoergendorfer.com/post/torchdcm-a-unified-pytorch-native-package-for-discrete-choice-modelingSecFeedde2026-08-21T07:16:24.318229+00:00TorchDCM: A Unified PyTorch-Native Package for Discrete Choice Modelinghttps://blog.schoergendorfer.com/post/m3-a-state-event-generative-foundation-model-for-market-microstructure-dynamicsSecFeedde2026-08-21T07:16:24.298392+00:00M3: A State-Event Generative Foundation Model for Market Microstructure Dynamicshttps://blog.schoergendorfer.com/post/vqc-zti-variational-quantum-control-for-zero-trust-protection-of-the-tactile-intSecFeedde2026-08-21T07:16:24.277947+00:00VQC-ZTI: Variational Quantum Control for Zero Trust Protection of the Tactile Internethttps://blog.schoergendorfer.com/post/a-comparison-between-ceiling-mounted-fmcw-ir-uwb-and-wi-fi-radar-for-in-bedroomSecFeedde2026-08-21T07:16:24.258088+00:00A comparison between ceiling-mounted FMCW, IR-UWB and Wi-Fi radar for in-bedroom human activity monitoring and sleep interruption detectionhttps://blog.schoergendorfer.com/post/explainable-transformer-models-for-clinical-prediction-tasks-on-structured-electSecFeedde2026-08-21T07:16:24.237719+00:00Explainable Transformer Models for Clinical Prediction Tasks on Structured Electronic Health Recordshttps://blog.schoergendorfer.com/post/physical-support-confidence-sets-for-highly-coherent-dictionariesSecFeedde2026-08-21T07:16:24.217852+00:00Physical-Support Confidence Sets for Highly Coherent Dictionarieshttps://blog.schoergendorfer.com/post/dics-data-informed-centroid-splitting-for-decision-tree-classifiersSecFeedde2026-08-21T07:16:24.181582+00:00DICS: Data-Informed Centroid Splitting for Decision Tree Classifiershttps://blog.schoergendorfer.com/post/decoding-silent-reading-from-non-invasive-eegSecFeedde2026-08-21T07:16:24.163747+00:00Decoding silent reading from non-invasive EEGhttps://blog.schoergendorfer.com/post/exact-algebraic-computation-of-learning-coefficients-for-two-dimensional-singulaSecFeedde2026-08-21T07:16:24.142526+00:00Exact Algebraic Computation of Learning Coefficients for Two-Dimensional Singular Modelshttps://blog.schoergendorfer.com/post/evaluating-neural-cartographic-relief-shading-for-urban-environments-a-downtownSecFeedde2026-08-21T07:01:24.682786+00:00Evaluating Neural Cartographic Relief Shading for Urban Environments: A Downtown Calgary Study Using High-Resolution DEM and DSM Datahttps://blog.schoergendorfer.com/post/sae-xplainers-rule-based-feature-interpretation-for-extreme-earth-eventsSecFeedde2026-08-21T07:01:24.663017+00:00SAE-Xplainers: Rule-Based Feature Interpretation for Extreme Earth Eventshttps://blog.schoergendorfer.com/post/decovae-a-lightweight-interpretable-trend-seasonal-vae-framework-for-efficient-pSecFeedde2026-08-21T07:01:24.643860+00:00DecoVAE: a Lightweight Interpretable Trend-Seasonal VAE Framework for Efficient Probabilistic Time Series Forecastinghttps://blog.schoergendorfer.com/post/orthogonal-jepa-factorized-predictive-states-for-latent-world-modelsSecFeedde2026-08-21T07:01:24.623923+00:00Orthogonal JEPA: Factorized Predictive States for Latent World Modelshttps://blog.schoergendorfer.com/post/end-to-end-early-classification-of-time-series-in-non-stationary-environmentsSecFeedde2026-08-21T07:01:24.601953+00:00End-to-end Early Classification of Time Series in Non-Stationary Environmentshttps://blog.schoergendorfer.com/post/clast-context-aware-contrastive-vae-for-probabilistic-time-series-forecastingSecFeedde2026-08-21T07:01:24.576980+00:00CLaST: Context-aware Contrastive VAE for Probabilistic Time Series Forecastinghttps://blog.schoergendorfer.com/post/systematic-evaluation-of-tabpfn-ts-for-zero-shot-probabilistic-heat-load-forecasSecFeedde2026-08-21T07:01:24.552381+00:00Systematic Evaluation of TabPFN-TS for Zero-Shot Probabilistic Heat Load Forecasting in District Heating Networkshttps://blog.schoergendorfer.com/post/scale-aware-pretraining-of-time-series-foundation-models-via-multi-patch-token-aSecFeedde2026-08-21T07:01:24.531249+00:00Scale-Aware Pretraining of Time Series Foundation Models via Multi-Patch Token Alignment and Hybrid Maskinghttps://blog.schoergendorfer.com/post/g-mark-grounded-multi-agent-reasoning-for-cooperative-driving-via-knowledge-grapSecFeedde2026-08-21T07:01:24.493764+00:00G-MARK: Grounded Multi-Agent Reasoning for Cooperative Driving via Knowledge Graphshttps://blog.schoergendorfer.com/post/auditing-recorded-predictive-lead-service-line-classifications-against-physicalSecFeedde2026-08-21T07:01:24.471844+00:00Auditing Recorded Predictive Lead Service-Line Classifications Against Physical Verification: A Statewide Study of New Yorkhttps://blog.schoergendorfer.com/post/multi-source-wasserstein-distributionally-robust-graph-learningSecFeedde2026-08-21T07:01:24.447072+00:00Multi-Source Wasserstein Distributionally Robust Graph Learninghttps://blog.schoergendorfer.com/post/petaparameter-efficient-test-time-adaptation-for-virtual-screeningSecFeedde2026-08-21T07:01:24.423406+00:00PETA:Parameter-Efficient Test-Time Adaptation for Virtual Screeninghttps://blog.schoergendorfer.com/post/reliable-neural-collapse-approximation-for-open-world-test-time-adaptationSecFeedde2026-08-21T07:01:24.399665+00:00Reliable Neural Collapse Approximation for Open-World Test-Time Adaptationhttps://blog.schoergendorfer.com/post/online-test-time-adaptation-for-generalizable-dynamic-graph-anomaly-detectionSecFeedde2026-08-21T07:01:24.378794+00:00Online Test-Time Adaptation for Generalizable Dynamic Graph Anomaly Detectionhttps://blog.schoergendorfer.com/post/far-dpo-feasibility-aware-and-robust-direct-preference-optimization-for-cyclic-pSecFeedde2026-08-21T07:01:24.355326+00:00FAR-DPO: Feasibility-Aware and Robust Direct Preference Optimization for Cyclic Peptide Designhttps://blog.schoergendorfer.com/post/inadvertent-context-leakage-in-language-modelsSecFeedde2026-08-21T07:01:24.330018+00:00Inadvertent Context Leakage in Language Modelshttps://blog.schoergendorfer.com/post/answer-level-trust-selection-for-physical-vision-language-reasoningSecFeedde2026-08-21T07:01:24.308022+00:00Answer-Level Trust Selection for Physical Vision-Language Reasoninghttps://blog.schoergendorfer.com/post/sage-xgboost-spatially-augmented-graph-embeddings-machine-learning-framework-forSecFeedde2026-08-21T06:46:22.211621+00:00SAGE-XGBoost: Spatially Augmented Graph Embeddings--Machine Learning Framework for Natural Hazards Susceptibility Mapping under Data Scarcityhttps://blog.schoergendorfer.com/post/fleetsieve-decision-critical-profiling-for-slo-aware-llm-fleet-configurationSecFeedde2026-08-21T06:46:22.184394+00:00FleetSieve: Decision-Critical Profiling for SLO-Aware LLM Fleet Configurationhttps://blog.schoergendorfer.com/post/rationally-enriched-chebyshev-trunk-bases-for-deeponet-surrogates-of-high-pecletSecFeedde2026-08-21T06:46:22.160885+00:00Rationally Enriched Chebyshev Trunk Bases for DeepONet Surrogates of High P\'eclet Entrance Transporthttps://blog.schoergendorfer.com/post/time-uniform-self-normalized-concentration-for-discounted-least-squares-limits-aSecFeedde2026-08-21T06:46:22.136174+00:00Time-Uniform Self-Normalized Concentration for Discounted Least Squares: Limits and Correctionshttps://blog.schoergendorfer.com/post/kahler-landscapes-for-complex-neural-network-descents-and-guarantees-including-aSecFeedde2026-08-21T06:46:22.088834+00:00K\"ahler landscapes for complex neural network descents and guarantees including a search and destroy of the Calabi-Yau manifoldhttps://blog.schoergendorfer.com/post/unregularized-convergence-of-single-loop-entropy-regularized-natural-actor-critiSecFeedde2026-08-21T06:46:22.059920+00:00Unregularized Convergence of Single-Loop, Entropy-Regularized Natural Actor-Critichttps://blog.schoergendorfer.com/post/a-two-stage-time-aware-transformer-for-short-horizon-aecopd-risk-predictionSecFeedde2026-08-21T06:46:22.031221+00:00A Two-Stage Time-Aware Transformer for Short-Horizon AECOPD Risk Predictionhttps://blog.schoergendorfer.com/post/continuous-adversarial-meanflow-transferSecFeedde2026-08-21T06:46:22.000391+00:00Continuous Adversarial MeanFlow Transferhttps://blog.schoergendorfer.com/post/empirical-characterization-of-learning-geometry-in-hybrid-quantum-forecasting-moSecFeedde2026-08-21T06:46:21.976287+00:00Empirical Characterization of Learning Geometry in Hybrid Quantum Forecasting Modelshttps://blog.schoergendorfer.com/post/beyond-multimodal-alignment-certifying-physical-language-through-response-substiSecFeedde2026-08-21T06:46:21.950920+00:00Beyond Multimodal Alignment: Certifying Physical Language through Response Substitution and Ordered Executionhttps://blog.schoergendorfer.com/post/when-to-retrain-an-empirical-study-of-retraining-policies-for-streaming-ml-underSecFeedde2026-08-21T06:46:21.924788+00:00When to Retrain: An Empirical Study of Retraining Policies for Streaming ML Under Concept Drift, Budget, and Latency Constraintshttps://blog.schoergendorfer.com/post/deltamomentum-a-key-value-based-anisotropic-momentum-update-via-delta-ruleSecFeedde2026-08-21T06:46:21.901678+00:00DeltaMomentum: A Key-Value based Anisotropic Momentum Update via Delta Rulehttps://blog.schoergendorfer.com/post/llm-as-detector-an-in-context-learning-approach-for-tabular-anomaly-detectionSecFeedde2026-08-21T06:46:21.876562+00:00LLM as Detector: An In-context Learning Approach for Tabular Anomaly Detectionhttps://blog.schoergendorfer.com/post/quantifying-event-impacts-on-time-series-via-multiscale-contrastive-learningSecFeedde2026-08-21T06:46:21.846478+00:00Quantifying Event Impacts on Time Series via Multiscale Contrastive Learninghttps://blog.schoergendorfer.com/post/uncovering-the-limits-of-proof-sharing-for-neural-networksSecFeedde2026-08-21T06:46:21.816647+00:00Uncovering the Limits of Proof Sharing for Neural Networkshttps://blog.schoergendorfer.com/post/holtercare-bench-a-multimodal-benchmark-for-evaluating-long-term-dynamic-ecg-anaSecFeedde2026-08-21T06:46:21.790766+00:00Holtercare-Bench: A Multimodal Benchmark for Evaluating Long-Term Dynamic ECG Analysishttps://blog.schoergendorfer.com/post/triangular-fuzzy-rescaling-distanceSecFeedde2026-08-21T06:46:21.766544+00:00Triangular Fuzzy Rescaling Distancehttps://blog.schoergendorfer.com/post/da-wam-decision-aligned-future-latents-for-driving-world-models-2SecFeedde2026-08-21T06:46:21.740507+00:00DA-WAM: Decision-Aligned Future Latents for Driving World Modelshttps://blog.schoergendorfer.com/post/graphical-design-of-interpretable-architectures-2SecFeedde2026-08-21T06:46:21.715647+00:00Graphical Design of Interpretable Architectureshttps://blog.schoergendorfer.com/post/too-sure-to-be-safe-model-calibration-for-reliable-log-anomaly-detection-2SecFeedde2026-08-21T06:31:28.912122+00:00Too Sure to Be Safe: Model Calibration for Reliable Log Anomaly Detectionhttps://blog.schoergendorfer.com/post/when-is-shallow-enough-adaptive-split-federated-learning-with-client-specific-su-2SecFeedde2026-08-21T06:31:28.849397+00:00When Is Shallow Enough? Adaptive Split Federated Learning with Client-Specific Sufficiency Estimationhttps://blog.schoergendorfer.com/post/why-do-ai-agents-break-rules-how-framing-context-and-social-signals-shape-compli-2SecFeedde2026-08-21T06:31:28.828162+00:00Why Do AI Agents Break Rules? How Framing, Context, and Social Signals Shape Compliancehttps://blog.schoergendorfer.com/post/socioduality-a-relational-process-framework-for-human-ai-interaction-2SecFeedde2026-08-21T06:31:28.806038+00:00Socioduality: A Relational Process Framework for Human-AI Interactionhttps://blog.schoergendorfer.com/post/a-model-internal-protocol-for-assessing-multimodal-models-as-integrated-systemsSecFeedde2026-08-21T06:31:28.784187+00:00A Model-Internal Protocol for Assessing Multimodal Models as Integrated Systemshttps://blog.schoergendorfer.com/post/finverse-financial-time-series-benchmark-2SecFeedde2026-08-21T06:31:28.759843+00:00FinVerse: Financial Time-Series Benchmarkhttps://blog.schoergendorfer.com/post/anatomy-contextualized-adaptation-of-ct-foundation-modelsSecFeedde2026-08-21T06:31:28.737617+00:00Anatomy Contextualized Adaptation of CT Foundation Modelshttps://blog.schoergendorfer.com/post/a-distributional-robustness-margin-for-pathology-foundation-models-2SecFeedde2026-08-21T06:31:28.715581+00:00A Distributional Robustness Margin For Pathology Foundation Modelshttps://blog.schoergendorfer.com/post/softvtbench-a-safety-aware-visuo-tactile-benchmark-for-physically-constrained-ro-2SecFeedde2026-08-21T06:31:28.691884+00:00SoftVTBench: A Safety-Aware Visuo-Tactile Benchmark for Physically Constrained Robotic Manipulation of Deformable Objects (Early Version)https://blog.schoergendorfer.com/post/drift-adaptive-icu-intervention-prediction-freezing-the-physiological-encoder-foSecFeedde2026-08-21T06:31:28.669477+00:00Drift-Adaptive ICU Intervention Prediction: Freezing the Physiological Encoder for Auditable Model Updatinghttps://blog.schoergendorfer.com/post/every-step-of-the-way-video-based-parkinsonian-turning-step-countingSecFeedde2026-08-21T06:31:28.645601+00:00Every Step of the Way: Video-based Parkinsonian Turning Step Countinghttps://blog.schoergendorfer.com/post/po-pddl-learning-symbolic-pomdps-from-visual-demonstrations-for-robot-planning-uSecFeedde2026-08-21T06:31:28.623748+00:00PO-PDDL: Learning Symbolic POMDPs from Visual Demonstrations for Robot Planning Under Uncertaintyhttps://blog.schoergendorfer.com/post/a-tool-to-map-ai-programs-in-the-us-a-snapshot-from-april-2026-and-an-analysis-oSecFeedde2026-08-21T06:31:28.602049+00:00A Tool to Map AI Programs in the U.S.: A Snapshot from April 2026 and an Analysis of Requirements for AI Majors and Minorshttps://blog.schoergendorfer.com/post/right-family-wrong-skill-benchmarking-risk-exposure-in-agent-skill-retrievalSecFeedde2026-08-21T06:31:28.562940+00:00Right Family, Wrong Skill: Benchmarking Risk Exposure in Agent Skill Retrievalhttps://blog.schoergendorfer.com/post/the-good-the-bad-and-the-ugly-of-markov-boundary-for-tabular-predictionSecFeedde2026-08-21T06:31:28.500072+00:00The Good, the Bad, and the Ugly of Markov Boundary for Tabular Predictionhttps://blog.schoergendorfer.com/post/beyond-recall-behavioral-specification-as-an-interpretive-layer-for-ai-personaliSecFeedde2026-08-21T06:16:22.313103+00:00Beyond Recall: Behavioral Specification as an Interpretive Layer for AI Personalizationhttps://blog.schoergendorfer.com/post/investigating-the-interplay-between-contextual-and-parametric-chain-of-thought-fSecFeedde2026-08-21T06:16:22.289822+00:00Investigating the Interplay between Contextual and Parametric Chain-of-Thought Faithfulness under Optimizationhttps://blog.schoergendorfer.com/post/geometric-evolution-maps-extracting-stable-concept-probes-from-transformer-residSecFeedde2026-08-21T06:16:22.266129+00:00Geometric Evolution Maps: Extracting Stable Concept Probes from Transformer Residual Streamshttps://blog.schoergendorfer.com/post/the-concept-allocation-zone-tracking-how-concepts-form-across-transformer-depthSecFeedde2026-08-21T06:16:22.238793+00:00The Concept Allocation Zone: Tracking How Concepts Form Across Transformer Depthhttps://blog.schoergendorfer.com/post/mkg-care-case-aware-reasoning-with-multimodal-knowledge-graphs-for-explainable-mSecFeedde2026-08-21T06:16:22.213157+00:00MKG-CARE: Case-Aware Reasoning with Multimodal Knowledge Graphs for Explainable Medical Image Diagnosishttps://blog.schoergendorfer.com/post/generative-ai-use-in-entrepreneurship-an-integrative-review-and-an-empowerment-eSecFeedde2026-08-21T06:16:22.187171+00:00Generative AI Use in Entrepreneurship: An Integrative Review and an Empowerment-Entrapment Frameworkhttps://blog.schoergendorfer.com/post/visd-enhancing-video-reasoning-via-structured-self-distillationSecFeedde2026-08-21T06:16:22.165830+00:00VISD: Enhancing Video Reasoning via Structured Self-Distillationhttps://blog.schoergendorfer.com/post/power-couple-ai-growth-and-renewable-energy-investmentSecFeedde2026-08-21T06:16:22.138705+00:00Power Couple? AI Growth and Renewable Energy Investmenthttps://blog.schoergendorfer.com/post/qworld-question-specific-evaluation-criteria-for-llmsSecFeedde2026-08-21T06:16:22.114518+00:00Qworld: Question-Specific Evaluation Criteria for LLMshttps://blog.schoergendorfer.com/post/extended-to-reality-prompt-injection-in-3d-environmentsSecFeedde2026-08-21T06:16:22.092301+00:00Extended to Reality: Prompt Injection in 3D Environmentshttps://blog.schoergendorfer.com/post/mambax-net-dual-input-mamba-enhanced-cross-attention-network-for-longitudinal-prSecFeedde2026-08-21T06:16:21.995706+00:00MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal Prostate MRI Segmentationhttps://blog.schoergendorfer.com/post/sum-agrivln-spatial-understanding-memory-for-agricultural-vision-and-language-na-2SecFeedde2026-08-21T06:16:21.976298+00:00SUM-AgriVLN: Spatial Understanding Memory for Agricultural Vision-and-Language Navigationhttps://blog.schoergendorfer.com/post/the-bidding-games-reinforcement-learning-for-mev-extraction-on-polygon-blockchaiSecFeedde2026-08-21T06:16:21.951059+00:00The Bidding Games: Reinforcement Learning for MEV Extraction on Polygon Blockchainhttps://blog.schoergendorfer.com/post/ts-reasoner-aligning-time-series-foundation-models-with-llm-reasoning-2SecFeedde2026-08-21T06:16:21.928996+00:00TS-Reasoner: Aligning Time Series Foundation Models with LLM Reasoninghttps://blog.schoergendorfer.com/post/fmtmathrmx-lazy-wavefront-search-for-dynamic-replanningSecFeedde2026-08-21T06:16:21.907000+00:00FMT$^{\mathrm{X}}$: Lazy Wavefront Search for Dynamic Replanninghttps://blog.schoergendorfer.com/post/the-basic-b-effect-the-use-of-llm-based-agents-reduces-the-distinctiveness-and-dSecFeedde2026-08-21T06:01:33.979274+00:00The Basic B*** Effect: The Use of LLM-based Agents Reduces the Distinctiveness and Diversity of People's Choiceshttps://blog.schoergendorfer.com/post/prefixagent-an-llm-powered-design-framework-for-efficient-prefix-adder-optimizatSecFeedde2026-08-21T06:01:33.956072+00:00PrefixAgent: An LLM-Powered Design Framework for Efficient Prefix Adder Optimizationhttps://blog.schoergendorfer.com/post/the-thousand-brains-theory-20-an-extension-for-the-long-range-connections-of-theSecFeedde2026-08-21T06:01:33.931548+00:00The Thousand Brains Theory 2.0: An Extension for the Long-Range Connections of the Neocortical Heterarchyhttps://blog.schoergendorfer.com/post/learning-augmented-power-system-operations-a-unified-optimization-viewSecFeedde2026-08-21T06:01:33.905387+00:00Learning-Augmented Power System Operations: A Unified Optimization Viewhttps://blog.schoergendorfer.com/post/regressor-guided-image-editing-shifts-emotion-and-disengagement-timing-in-socialSecFeedde2026-08-21T06:01:33.840774+00:00Regressor-Guided Image Editing Shifts Emotion and Disengagement Timing in Social Mediahttps://blog.schoergendorfer.com/post/your-turn-at-home-turning-angle-estimation-for-parkinsons-disease-severity-assesSecFeedde2026-08-21T06:01:33.799882+00:00Your Turn: At Home Turning Angle Estimation for Parkinson's Disease Severity Assessmenthttps://blog.schoergendorfer.com/post/virtual-sensing-to-enable-real-time-monitoring-of-inaccessible-locations-unmeasuSecFeedde2026-08-21T06:01:33.777711+00:00Virtual Sensing to Enable Real-Time Monitoring of Inaccessible Locations & Unmeasurable Parametershttps://blog.schoergendorfer.com/post/towards-efficient-pareto-set-approximation-via-mixture-of-experts-based-model-fuSecFeedde2026-08-21T06:01:33.756322+00:00Towards Efficient Pareto Set Approximation via Mixture of Experts Based Model Fusionhttps://blog.schoergendorfer.com/post/teacher-free-latent-self-distillation-and-class-separable-representations-for-liSecFeedde2026-08-21T06:01:33.731942+00:00Teacher-free Latent Self-distillation and Class-separable Representations for Lightweight IoT Attack Detectionhttps://blog.schoergendorfer.com/post/learning-based-speed-estimation-from-accelerometer-only-inertial-sensingSecFeedde2026-08-21T06:01:33.703219+00:00Learning-Based Speed Estimation from Accelerometer-Only Inertial Sensinghttps://blog.schoergendorfer.com/post/testifai-tomography-based-testing-for-deep-learning-systemsSecFeedde2026-08-21T06:01:33.683099+00:00TestifAI: Tomography-Based Testing for Deep Learning Systemshttps://blog.schoergendorfer.com/post/which-negatives-matter-ask-your-text-encoder-adaptive-similarity-margins-for-den-2SecFeedde2026-08-21T06:01:33.663646+00:00Which Negatives Matter? Ask Your Text Encoder: Adaptive Similarity Margins for Dense-Caption Retrievalhttps://blog.schoergendorfer.com/post/graph-surgery-and-the-do-operator-a-precise-correspondence-for-acyclic-structura-2SecFeedde2026-08-21T06:01:33.642465+00:00Graph Surgery and the Do-Operator: A Precise Correspondence for Acyclic Structural Causal Modelshttps://blog.schoergendorfer.com/post/longrca-bench-diagnosing-responsible-roles-and-root-causes-in-long-horizon-agent-2SecFeedde2026-08-21T06:01:33.619928+00:00LongRCA Bench: Diagnosing Responsible Roles and Root Causes in Long-Horizon Agent Failureshttps://blog.schoergendorfer.com/post/andy-a-mathematical-agent-for-rigorous-proof-and-autonomous-research-2SecFeedde2026-08-21T06:01:33.597231+00:00Andy: A Mathematical Agent for Rigorous Proof and Autonomous Researchhttps://blog.schoergendorfer.com/post/genco-a-unified-neural-solver-embedded-in-a-development-framework-for-steady-sta-2SecFeedde2026-08-21T06:01:33.576008+00:00GENCO - A Unified Neural Solver Embedded in a Development Framework for Steady-State Grid Analysishttps://blog.schoergendorfer.com/post/wnm-3d-a-world-navigation-model-with-3d-scene-conditioning-for-closed-loop-vln-2SecFeedde2026-08-21T06:01:33.555497+00:00WNM-3D: A World Navigation Model with 3D Scene Conditioning for Closed-Loop VLNhttps://blog.schoergendorfer.com/post/evaluating-investment-logic-in-large-language-models-a-real-world-benchmark-towa-2SecFeedde2026-08-21T05:46:17.613472+00:00Evaluating Investment Logic in Large Language Models: A Real-World Benchmark Towards Personalzied Financial Agentshttps://blog.schoergendorfer.com/post/llm-capability-limits-static-emergence-and-dynamic-boundary-controlSecFeedde2026-08-21T05:46:17.592556+00:00LLM Capability Limits: Static Emergence and Dynamic Boundary Controlhttps://blog.schoergendorfer.com/post/flowevo-self-evolving-agents-through-the-co-evolution-of-workflows-and-executablSecFeedde2026-08-21T05:46:17.570665+00:00FlowEvo: Self-Evolving Agents through the Co-Evolution of Workflows and Executable Skillshttps://blog.schoergendorfer.com/post/cadre-stable-parameter-efficient-adaptation-of-medical-vision-language-models-wiSecFeedde2026-08-21T05:46:17.549333+00:00CADRE: Stable, Parameter Efficient Adaptation of Medical Vision Language Models with Bounded Forgetting and Prior Drifthttps://blog.schoergendorfer.com/post/from-accuracy-to-auditability-a-survey-of-determinism-in-financial-ai-systemsSecFeedde2026-08-21T05:46:17.526422+00:00From Accuracy to Auditability: A Survey of Determinism in Financial AI Systemshttps://blog.schoergendorfer.com/post/chronoagentic-a-code-based-multi-agent-world-simulator-for-physically-grounded-sSecFeedde2026-08-21T05:46:17.502567+00:00ChronoAgentic: A Code-based Multi-Agent World Simulator for Physically Grounded Simulation Constructionhttps://blog.schoergendorfer.com/post/the-first-drop-of-ink-nonlinear-impact-of-distracting-information-in-long-contexSecFeedde2026-08-21T05:46:17.478034+00:00The First Drop of Ink: Nonlinear Impact of Distracting Information in Long-Context Reasoninghttps://blog.schoergendorfer.com/post/agent-first-tool-api-a-semantic-interface-paradigm-for-enterprise-ai-agent-systeSecFeedde2026-08-21T05:46:17.457621+00:00Agent-First Tool API: A Semantic Interface Paradigm for Enterprise AI Agent Systemshttps://blog.schoergendorfer.com/post/every-picture-tells-a-dangerous-story-memory-augmented-multi-agent-jailbreak-att-2SecFeedde2026-08-21T05:46:17.436539+00:00Every Picture Tells a Dangerous Story: Memory-Augmented Multi-Agent Jailbreak Attacks on VLMshttps://blog.schoergendorfer.com/post/chartool-tool-integrated-visual-reasoning-for-chart-understandingSecFeedde2026-08-21T05:46:17.415737+00:00CharTool: Tool-Integrated Visual Reasoning for Chart Understandinghttps://blog.schoergendorfer.com/post/atbench-a-diverse-and-realistic-agent-trajectory-benchmark-for-safety-evaluationSecFeedde2026-08-21T05:46:17.395082+00:00ATBench: A Diverse and Realistic Agent Trajectory Benchmark for Safety Evaluation and Diagnosishttps://blog.schoergendorfer.com/post/computational-phenomenology-of-borderline-personality-disorder-a-comparative-evaSecFeedde2026-08-21T05:46:17.358925+00:00Computational Phenomenology of Borderline Personality Disorder: A Comparative Evaluation of LLM-Simulated Expert Personas and Human Clinical Expertshttps://blog.schoergendorfer.com/post/toward-greater-autonomy-in-materials-discovery-agents-unifying-planning-physicsSecFeedde2026-08-21T05:46:17.317739+00:00Toward Greater Autonomy in Materials Discovery Agents: Unifying Planning, Physics, and Scientistshttps://blog.schoergendorfer.com/post/g-carl-grounded-checklist-aligned-reward-learning-for-patient-oriented-medical-rSecFeedde2026-08-21T05:46:17.297758+00:00G-CARL: Grounded Checklist-Aligned Reward Learning for Patient-Oriented Medical Report Interpretationhttps://blog.schoergendorfer.com/post/inject-align-recover-staged-post-training-for-retrieval-free-document-knowledgeSecFeedde2026-08-21T05:46:17.274777+00:00Inject, Align, Recover: Staged Post-Training for Retrieval-Free Document Knowledge Internalizationhttps://blog.schoergendorfer.com/post/inducing-task-models-from-computer-use-tracesSecFeedde2026-08-21T05:46:17.249398+00:00Inducing Task Models from Computer-Use Traceshttps://blog.schoergendorfer.com/post/growth-without-us-machine-consumers-corporate-circularity-and-the-decoupling-ofSecFeedde2026-08-21T05:46:17.224105+00:00Growth Without Us: Machine Consumers, Corporate Circularity, and the Decoupling of GDP from Humanity after AGIhttps://blog.schoergendorfer.com/post/prompt-conditioned-channel-attention-for-hierarchical-feature-modulation-towardSecFeedde2026-08-21T05:46:17.198200+00:00Prompt-Conditioned Channel Attention for Hierarchical Feature Modulation toward Anatomy-Agnostic Segmentationhttps://blog.schoergendorfer.com/post/daedalus-150m-a-convolution-attention-hybrid-designed-for-cpu-inferenceSecFeedde2026-08-21T05:31:18.599463+00:00Daedalus-150M: A Convolution-Attention Hybrid Designed for CPU Inferencehttps://blog.schoergendorfer.com/post/from-agent-behaviour-to-agent-friendly-documentation-an-empirical-study-of-how-cSecFeedde2026-08-21T05:31:18.577230+00:00From Agent Behaviour to Agent-Friendly Documentation: An Empirical Study of How Coding Agents Discover, Read, and Write Technical Documentationhttps://blog.schoergendorfer.com/post/multi-method-causal-evidence-synthesis-ranking-candidate-drivers-by-convergent-cSecFeedde2026-08-21T05:31:18.557441+00:00Multi-Method Causal Evidence Synthesis: Ranking Candidate Drivers by Convergent Cross-Method Evidence from Observational Datahttps://blog.schoergendorfer.com/post/a-standardized-framework-for-machine-learning-in-power-system-protectionSecFeedde2026-08-21T05:31:18.536918+00:00A Standardized Framework for Machine Learning in Power System Protectionhttps://blog.schoergendorfer.com/post/task-coevolve-efficient-harness-optimization-via-adaptive-validation-task-selectSecFeedde2026-08-21T05:31:18.516116+00:00Task-CoEvolve: Efficient Harness Optimization via Adaptive Validation Task Selectionhttps://blog.schoergendorfer.com/post/structured-affinity-for-unsupervised-visual-class-incremental-memory-in-deep-artSecFeedde2026-08-21T05:31:18.495535+00:00Structured Affinity for Unsupervised Visual Class-Incremental Memory in Deep Artificial Immune Networkshttps://blog.schoergendorfer.com/post/towards-professional-tennis-styles-for-humanoid-robots-with-adaptive-motion-planSecFeedde2026-08-21T05:31:18.472266+00:00Towards Professional Tennis Styles for Humanoid Robots with Adaptive Motion Planning and Trackinghttps://blog.schoergendorfer.com/post/evidence-gated-task-and-motion-planning-with-vision-language-modelsSecFeedde2026-08-21T05:31:18.452839+00:00Evidence-Gated Task and Motion Planning with Vision-Language Modelshttps://blog.schoergendorfer.com/post/lets-scale-step-by-step-compute-efficient-hyperparameter-transfer-for-large-scalSecFeedde2026-08-21T05:31:18.414987+00:00Let's Scale Step by Step: Compute-Efficient Hyperparameter Transfer for Large-Scale Mixture-of-Expertshttps://blog.schoergendorfer.com/post/echocot-extracting-hidden-chain-of-thought-from-large-reasoning-modelsSecFeedde2026-08-21T05:31:18.391259+00:00EchoCoT: Extracting Hidden Chain-of-Thought from Large Reasoning Modelshttps://blog.schoergendorfer.com/post/open-vocabulary-3d-object-detection-with-co-distillation-discovery-and-dual-guidSecFeedde2026-08-21T05:31:18.368561+00:00Open-Vocabulary 3D Object Detection with Co-Distillation Discovery and Dual Guidance Robust Traininghttps://blog.schoergendorfer.com/post/an-inclusive-and-lightweight-approach-to-federated-continual-learning-for-culturSecFeedde2026-08-21T05:31:18.346689+00:00An Inclusive and Lightweight Approach to Federated Continual Learning for Cultural Heritagehttps://blog.schoergendorfer.com/post/designing-human-mediated-ai-guidance-ready-together-for-personalized-family-emerSecFeedde2026-08-21T05:31:18.324320+00:00Designing Human-mediated AI Guidance: Ready Together for Personalized Family Emergency Preparednesshttps://blog.schoergendorfer.com/post/towards-quantifying-benchmark-optimization-in-asr-modelsSecFeedde2026-08-21T05:31:18.301493+00:00Towards Quantifying Benchmark Optimization in ASR Modelshttps://blog.schoergendorfer.com/post/maliciousskillbench-a-comprehensive-benchmark-for-malicious-agent-skill-detectioSecFeedde2026-08-21T05:31:18.279829+00:00MaliciousSkillBench: A Comprehensive Benchmark for Malicious Agent Skill Detectionhttps://blog.schoergendorfer.com/post/interrupting-the-loop-periodic-subject-changes-raise-judged-surprise-and-connectSecFeedde2026-08-21T05:31:18.256590+00:00Interrupting the Loop: Periodic Subject Changes Raise Judged Surprise and Connection in Base Language Modelshttps://blog.schoergendorfer.com/post/evidence-before-expansion-reuse-spawn-or-defer-in-lifelong-expert-poolsSecFeedde2026-08-21T05:31:18.235607+00:00Evidence Before Expansion: Reuse, Spawn, or Defer in Lifelong Expert Poolshttps://blog.schoergendorfer.com/post/separating-covariate-shift-from-mechanism-change-with-two-discriminators-cjsd-aSecFeedde2026-08-21T05:31:18.212580+00:00Separating Covariate Shift from Mechanism Change with Two Discriminators: CJSD, a Conditional Discrepancy with an Exact Covariate-Concept Decompositionhttps://blog.schoergendorfer.com/post/a-knowledge-guided-agentic-framework-for-mitigating-patient-context-ambiguity-inSecFeedde2026-08-21T05:31:18.187208+00:00A knowledge-guided agentic framework for mitigating patient-context ambiguity in health querieshttps://blog.schoergendorfer.com/post/listening-forward-next-patch-embedding-prediction-enables-scalable-audio-learnerSecFeedde2026-08-21T05:16:17.532109+00:00Listening Forward: Next Patch Embedding Prediction Enables Scalable Audio Learnershttps://blog.schoergendorfer.com/post/adaptive-probabilistic-shielding-by-learning-mdps-for-safe-reinforcement-learninSecFeedde2026-08-21T05:16:17.506582+00:00Adaptive Probabilistic Shielding by Learning MDPs for Safe Reinforcement Learninghttps://blog.schoergendorfer.com/post/repo0-design-driven-zero-to-all-code-generationSecFeedde2026-08-21T05:16:17.481438+00:00Repo0: Design-Driven Zero-to-All Code Generationhttps://blog.schoergendorfer.com/post/core-kan-continuous-vision-kernels-with-kolmogorov-arnold-networksSecFeedde2026-08-21T05:16:17.458475+00:00Core-KAN: Continuous Vision Kernels with Kolmogorov-Arnold Networkshttps://blog.schoergendorfer.com/post/milegpo-milestone-inference-with-local-evidence-for-graph-based-policy-optimizatSecFeedde2026-08-21T05:16:17.431847+00:00MileGPO: Milestone Inference with Local Evidence for Graph-Based Policy Optimization of Long-Horizon LLM Agentshttps://blog.schoergendorfer.com/post/lora-ga2-low-rank-adaptation-with-multi-step-gradient-adaptive-alignmentSecFeedde2026-08-21T05:16:17.407721+00:00LoRA-GA$^2$: Low Rank Adaptation with Multi-step Gradient Adaptive Alignmenthttps://blog.schoergendorfer.com/post/an-irreducible-quantum-advantage-in-aligning-world-models-with-realitySecFeedde2026-08-21T05:16:17.383557+00:00An Irreducible Quantum Advantage in Aligning World Models with Realityhttps://blog.schoergendorfer.com/post/distilling-aggregate-mobility-statistics-into-a-language-model-policy-for-post-eSecFeedde2026-08-21T05:16:17.361829+00:00Distilling Aggregate Mobility Statistics into a Language Model Policy for Post-Event Crowd Simulationhttps://blog.schoergendorfer.com/post/cotograsp-contact-topology-conditioned-dexterous-grasp-synthesis-via-canonical-wSecFeedde2026-08-21T05:16:17.336420+00:00CoToGrasp: Contact-Topology-Conditioned Dexterous Grasp Synthesis via Canonical Workspace Learninghttps://blog.schoergendorfer.com/post/finite-horizon-input-output-dynamics-of-minibatch-perturbations-in-adamwSecFeedde2026-08-21T05:16:17.314616+00:00Finite-Horizon Input-Output Dynamics of Minibatch Perturbations in AdamWhttps://blog.schoergendorfer.com/post/credit-without-ground-truth-auditing-step-level-credit-assignment-in-llm-agentsSecFeedde2026-08-21T05:16:17.293265+00:00Credit Without Ground Truth: Auditing Step-Level Credit Assignment in LLM Agents Against Executed Replayhttps://blog.schoergendorfer.com/post/goag-generative-and-object-agnostic-grasp-planner-for-dexterous-robotic-manipulaSecFeedde2026-08-21T05:16:17.269674+00:00GOAG: Generative and Object-Agnostic Grasp Planner for Dexterous Robotic Manipulationhttps://blog.schoergendorfer.com/post/truncate-bad-upweight-good-bon-style-distillation-via-rank-based-classificationSecFeedde2026-08-21T05:16:17.247426+00:00Truncate Bad, Upweight Good: BoN-Style Distillation via Rank-Based Classificationhttps://blog.schoergendorfer.com/post/question-guided-evidence-acquisition-for-multimodal-visual-question-answeringSecFeedde2026-08-21T05:16:17.222284+00:00Question-Guided Evidence Acquisition for Multimodal Visual Question Answeringhttps://blog.schoergendorfer.com/post/learning-to-beat-phenotype-guided-latent-flow-with-regional-motion-priors-for-biSecFeedde2026-08-21T05:16:17.201163+00:00Learning to Beat: Phenotype-Guided Latent Flow with Regional Motion Priors for Biventricular Motion Synthesishttps://blog.schoergendorfer.com/post/tempjail-temporal-jailbreak-attack-against-large-vision-language-models-via-subtSecFeedde2026-08-21T05:16:17.179712+00:00TempJail: Temporal Jailbreak Attack against Large Vision-Language Models via Subtitle Schedulinghttps://blog.schoergendorfer.com/post/a-locally-tokenized-generative-model-for-robust-time-series-watermarkingSecFeedde2026-08-21T05:16:17.157478+00:00A Locally Tokenized Generative Model for Robust Time-Series Watermarkinghttps://blog.schoergendorfer.com/post/scale-separated-conditioning-for-style-encoder-free-diffusion-stylizationSecFeedde2026-08-21T05:16:17.133124+00:00Scale-Separated Conditioning for Style-Encoder-Free Diffusion Stylizationhttps://blog.schoergendorfer.com/post/robust-cross-modal-foundation-model-perception-for-underwater-robots-under-degraSecFeedde2026-08-21T05:16:17.112027+00:00Robust Cross-Modal Foundation Model Perception for Underwater Robots under Degraded Visual Conditionshttps://blog.schoergendorfer.com/post/loreley-repository-scale-program-evolution-with-quality-diversity-searchSecFeedde2026-08-21T05:16:17.085697+00:00Loreley: Repository-Scale Program Evolution with Quality-Diversity Searchhttps://blog.schoergendorfer.com/post/escaping-the-quicksand-a-call-to-armsSecFeedde2026-08-21T05:01:32.914397+00:00Escaping the Quicksand: A Call to Armshttps://blog.schoergendorfer.com/post/forking-fast-efficiently-estimating-uncertainty-dynamics-in-text-generationSecFeedde2026-08-21T05:01:32.891120+00:00Forking Fast: Efficiently Estimating Uncertainty Dynamics in Text Generationhttps://blog.schoergendorfer.com/post/pea-dpo-perception-enhanced-alignment-direct-preference-optimization-for-mllms-aSecFeedde2026-08-21T05:01:32.854526+00:00PEA-DPO: Perception-Enhanced Alignment Direct Preference Optimization for MLLMs Alignmenthttps://blog.schoergendorfer.com/post/vgi-bench-probing-visual-intelligence-in-video-generation-modelsSecFeedde2026-08-21T05:01:32.833777+00:00VGI-BENCH: Probing Visual Intelligence in Video Generation Modelshttps://blog.schoergendorfer.com/post/draftfm-a-foundationmodel-for-day-zero-drafting-in-magic-the-gatheringSecFeedde2026-08-21T05:01:32.812255+00:00DraftFM: A FoundationModel for Day-Zero Drafting in Magic: The Gatheringhttps://blog.schoergendorfer.com/post/cvsd-reg-cross-modal-visual-semantic-prior-distillation-for-robust-lidar-registrSecFeedde2026-08-21T05:01:32.793020+00:00CVSD-Reg: Cross-Modal Visual Semantic Prior Distillation for Robust LiDAR Registrationhttps://blog.schoergendorfer.com/post/stream4d-4d-consistency-for-streaming-autoregressive-diffusion-video-modelsSecFeedde2026-08-21T05:01:32.770838+00:00Stream4D: 4D-Consistency for Streaming Autoregressive Diffusion Video Modelshttps://blog.schoergendorfer.com/post/when-machines-speak-a-unified-generative-framework-for-integrating-machine-nativSecFeedde2026-08-21T05:01:32.748314+00:00When Machines Speak: A Unified Generative Framework for Integrating Machine-Native Symbols into Pretrained Large Language Modelshttps://blog.schoergendorfer.com/post/automated-summarization-of-financial-news-using-large-language-models-and-retrieSecFeedde2026-08-21T05:01:32.727775+00:00Automated Summarization of Financial News Using Large Language Models and Retrieval-Augmented Generation: An Early Empirical Study (Fall 2023)https://blog.schoergendorfer.com/post/in-two-minds-about-lifelong-learning-exploring-hemispheric-redundancy-and-speciaSecFeedde2026-08-21T05:01:32.706812+00:00In Two Minds about Lifelong Learning: Exploring Hemispheric Redundancy and Specialisation in Neural Modelshttps://blog.schoergendorfer.com/post/measuring-what-a-specification-determines-a-formal-semantic-block-model-and-an-eSecFeedde2026-08-21T05:01:32.687293+00:00Measuring What a Specification Determines: A Formal Semantic-Block Model and an Execution-Judged Benchmarkhttps://blog.schoergendorfer.com/post/accelerated-genetic-programming-hyper-heuristics-for-simulation-based-schedulingSecFeedde2026-08-21T05:01:32.667807+00:00Accelerated Genetic Programming Hyper-Heuristics for Simulation-Based Scheduling via Agentic AIhttps://blog.schoergendorfer.com/post/are-llms-becoming-similarly-creative-evidence-from-three-years-of-modelsSecFeedde2026-08-21T05:01:32.646298+00:00Are LLMs becoming similarly creative? Evidence from three years of modelshttps://blog.schoergendorfer.com/post/longitudinal-bayesian-learning-of-continuous-disease-position-across-the-alzheimSecFeedde2026-08-21T05:01:32.623244+00:00Longitudinal Bayesian Learning of Continuous Disease Position across the Alzheimer's Disease Continuumhttps://blog.schoergendorfer.com/post/scape-scenario-conditioned-simulation-augmented-policy-evaluationSecFeedde2026-08-21T05:01:32.599310+00:00SCAPE: Scenario-Conditioned Simulation-Augmented Policy Evaluationhttps://blog.schoergendorfer.com/post/hydra-a-heterogeneous-chiplet-dse-framework-for-serving-dynamic-hybrid-llm-worklSecFeedde2026-08-21T05:01:32.555986+00:00HYDRA: A Heterogeneous Chiplet DSE Framework for Serving Dynamic Hybrid LLM Workloadshttps://blog.schoergendorfer.com/post/fairness-aware-network-embeddings-methods-applications-and-challengesSecFeedde2026-08-21T05:01:32.518110+00:00Fairness-Aware Network Embeddings: Methods, Applications, and Challengeshttps://blog.schoergendorfer.com/post/does-marginal-coverage-guarantee-class-conditional-safety-for-zero-shot-vlms-undSecFeedde2026-08-21T04:46:25.995951+00:00Does Marginal Coverage Guarantee Class-Conditional Safety for Zero-Shot VLMs Under Shift?https://blog.schoergendorfer.com/post/mechanistic-tomography-designed-measurement-for-control-oriented-interpretabilitSecFeedde2026-08-21T04:46:25.974604+00:00Mechanistic Tomography: Designed Measurement for Control-Oriented Interpretabilityhttps://blog.schoergendorfer.com/post/quantum-kernel-estimation-for-the-discovery-of-early-lung-cancer-detectionSecFeedde2026-08-21T04:46:25.936244+00:00Quantum Kernel Estimation for the Discovery of Early Lung Cancer Detectionhttps://blog.schoergendorfer.com/post/mapping-general-purpose-ai-governance-in-twenty-ai-middle-power-jurisdictionsSecFeedde2026-08-21T04:46:25.910524+00:00Mapping General-Purpose AI Governance in Twenty AI Middle-Power Jurisdictionshttps://blog.schoergendorfer.com/post/incident-data-robustness-analysis-of-the-owasp-top-10-for-llm-applications-2026SecFeedde2026-08-21T04:46:25.881596+00:00Incident-Data Robustness Analysis of the OWASP Top 10 for LLM Applications (2026): How a Community-Expert Ranking Holds Up Against a Large-Scale LLM Incident Corpushttps://blog.schoergendorfer.com/post/active-spiking-perception-the-membrane-potential-as-a-belief-state-for-anytime-3SecFeedde2026-08-21T04:46:25.859477+00:00Active Spiking Perception: The Membrane Potential as a Belief State for Anytime 3D Point Cloud Recognitionhttps://blog.schoergendorfer.com/post/when-ai-writes-who-gets-cited-evidence-of-citation-monoculture-across-language-mSecFeedde2026-08-21T04:46:25.834338+00:00When AI Writes, Who Gets Cited? Evidence of Citation Monoculture Across Language Modelshttps://blog.schoergendorfer.com/post/causal-inference-under-interference-with-learned-exposure-mappingsSecFeedde2026-08-21T04:46:25.811100+00:00Causal Inference under Interference with Learned Exposure Mappingshttps://blog.schoergendorfer.com/post/can-conversational-ai-loosen-us-versus-them-boundaries-the-effects-of-common-duaSecFeedde2026-08-21T04:46:25.785120+00:00Can Conversational AI loosen Us-Versus-Them Boundaries? The Effects of Common, Dual, and Separate Identity Framings on Pro-Immigrant Intergroup Helpinghttps://blog.schoergendorfer.com/post/represented-but-ignored-a-causal-account-of-prosodic-underuse-in-audio-languageSecFeedde2026-08-21T04:46:25.759799+00:00Represented but Ignored: A Causal Account of Prosodic Underuse in Audio-Language Modelshttps://blog.schoergendorfer.com/post/time-series-retrieval-for-grounding-multimodal-language-models-in-remaining-usefSecFeedde2026-08-21T04:46:25.734953+00:00Time-Series Retrieval for Grounding Multimodal Language Models in Remaining Useful Lifehttps://blog.schoergendorfer.com/post/hallucination-as-a-feature-not-a-defect-evaluating-a-multi-agent-architecture-toSecFeedde2026-08-21T04:46:25.691772+00:00Hallucination as a Feature, not a Defect: Evaluating a multi-agent architecture to transform speculative language-model outputs into testable scientific hypotheseshttps://blog.schoergendorfer.com/post/asymmetric-attention-heads-structured-head-wise-context-allocation-for-transformSecFeedde2026-08-21T04:46:25.671871+00:00Asymmetric Attention Heads: Structured Head-Wise Context Allocation for Transformer Attentionhttps://blog.schoergendorfer.com/post/automatic-bioinformatic-software-named-entity-recognition-from-literatureSecFeedde2026-08-21T04:46:25.651744+00:00Automatic bioinformatic software named entity recognition from literaturehttps://blog.schoergendorfer.com/post/a-virtual-member-of-a-community-of-practice-for-the-society-of-petroleum-engineeSecFeedde2026-08-21T04:46:25.607938+00:00A Virtual Member of a Community of Practice for the Society of Petroleum Engineers: From Prototype to Deploymenthttps://blog.schoergendorfer.com/post/an-agentic-approach-for-active-data-collection-travel-behavior-modeling-and-weatSecFeedde2026-08-21T04:46:25.583333+00:00An Agentic Approach for Active Data Collection, Travel Behavior Modeling, and Weather-Sensitive Demand Predictionhttps://blog.schoergendorfer.com/post/ai4ai-bench-benchmarking-llm-agents-in-algorithmic-design-for-recursive-self-impSecFeedde2026-08-21T04:46:25.554732+00:00AI4AI-Bench: Benchmarking LLM Agents in Algorithmic Design for Recursive Self-Improvementhttps://blog.schoergendorfer.com/post/pandoras-ai-model-routing-box-efficient-allocation-with-costly-value-estimationSecFeedde2026-08-21T04:31:27.486246+00:00Pandora's AI Model Routing Box: Efficient Allocation with Costly Value Estimationhttps://blog.schoergendorfer.com/post/midtool-mid-training-data-synthesis-for-agentic-tool-useSecFeedde2026-08-21T04:31:27.459519+00:00MidTool: Mid-training Data Synthesis for Agentic Tool Usehttps://blog.schoergendorfer.com/post/phantom-gains-auditing-self-improvement-against-a-measured-nullSecFeedde2026-08-21T04:31:27.436727+00:00Phantom Gains: Auditing Self-Improvement Against a Measured Nullhttps://blog.schoergendorfer.com/post/catching-the-rug-early-prediction-of-fraudulent-memecoins-on-solana-via-machineSecFeedde2026-08-21T04:31:27.415786+00:00Catching the Rug: Early Prediction of Fraudulent Memecoins on Solana via Machine Learninghttps://blog.schoergendorfer.com/post/break-it-down-pass-it-on-cross-task-skill-transfer-in-llm-agentsSecFeedde2026-08-21T04:31:27.391400+00:00Break It Down, Pass It On: Cross-Task Skill Transfer in LLM Agentshttps://blog.schoergendorfer.com/post/learning-when-to-think-adaptive-reasoning-for-test-time-compute-allocationSecFeedde2026-08-21T04:31:27.366496+00:00Learning When to Think: Adaptive Reasoning for Test-Time Compute Allocationhttps://blog.schoergendorfer.com/post/quasar-a-quantum-classical-neural-network-for-sar-satellite-physical-layer-autheSecFeedde2026-08-21T04:31:27.347291+00:00QUASAR: A Quantum-Classical Neural Network for SAR Satellite Physical-Layer Authenticationhttps://blog.schoergendorfer.com/post/rule-compliant-visual-spatial-planning-for-multimodal-large-language-modelsSecFeedde2026-08-21T04:31:27.325693+00:00Rule-Compliant Visual Spatial Planning for Multimodal Large Language Modelshttps://blog.schoergendorfer.com/post/insufficiencybench-evaluating-llm-legal-advice-on-underspecified-user-queriesSecFeedde2026-08-21T04:31:27.301523+00:00InsufficiencyBench: Evaluating LLM legal advice on underspecified user querieshttps://blog.schoergendorfer.com/post/electronic-navigational-chart-change-classificationSecFeedde2026-08-21T04:31:27.272481+00:00Electronic Navigational Chart Change Classificationhttps://blog.schoergendorfer.com/post/contractscrub-a-benchmark-for-final-review-of-legal-contractsSecFeedde2026-08-21T04:31:27.246140+00:00ContractScrub: A benchmark for final review of legal contractshttps://blog.schoergendorfer.com/post/memtrapbench-benchmarking-cognitive-traps-in-llm-memory-useSecFeedde2026-08-21T04:31:27.219878+00:00MemTrapBench: Benchmarking Cognitive Traps in LLM Memory Usehttps://blog.schoergendorfer.com/post/the-third-restructuring-of-software-form-from-the-three-tier-architecture-to-stoSecFeedde2026-08-21T04:31:27.195356+00:00The Third Restructuring of Software Form: From the Three-Tier Architecture to Storage, Models, and Agentshttps://blog.schoergendorfer.com/post/dars-dual-level-credit-assignment-rl-with-structured-reasoning-for-instruction-bSecFeedde2026-08-21T04:31:27.170013+00:00DARS: Dual-Level Credit Assignment RL with Structured Reasoning for Instruction-Based Image Editinghttps://blog.schoergendorfer.com/post/decowam-decoupled-whole-body-world-action-model-for-legged-mobile-manipulationSecFeedde2026-08-21T04:31:27.144786+00:00DECOWAM: Decoupled Whole-Body World-Action Model for Legged Mobile Manipulationhttps://blog.schoergendorfer.com/post/what-you-cant-see-is-what-you-learn-restricted-evidence-visibility-favors-composSecFeedde2026-08-21T04:31:27.121258+00:00What You Can't See Is What You Learn: Restricted Evidence Visibility Favors Compositional Generalization in Shared-Genome Language-Model Societieshttps://blog.schoergendorfer.com/post/on-the-applicability-of-safety-nets-a-safety-by-design-solution-for-certifying-nSecFeedde2026-08-21T04:31:27.095327+00:00On the Applicability of Safety Nets: A Safety-By-Design Solution for Certifying Neural Networkshttps://blog.schoergendorfer.com/post/a-three-dimensional-typology-of-agency-for-advanced-ai-systemsSecFeedde2026-08-21T04:31:27.072468+00:00A three-dimensional typology of agency for advanced AI systemshttps://blog.schoergendorfer.com/post/contrastive-mixed-prompt-learning-for-incomplete-multimodal-sentiment-analysis-wSecFeedde2026-08-21T04:31:27.046798+00:00Contrastive Mixed Prompt Learning for Incomplete Multimodal Sentiment Analysis with Unseen Modality Combinationhttps://blog.schoergendorfer.com/post/manifold-drift-in-flow-preference-optimization-a-root-cause-of-reward-hackingSecFeedde2026-08-21T04:31:27.023515+00:00Manifold Drift in Flow Preference Optimization: A Root Cause of Reward Hackinghttps://blog.schoergendorfer.com/post/exphy-a-benchmark-for-explicit-physical-property-learning-in-multi-object-trajecSecFeedde2026-08-21T04:16:30.060398+00:00ExPhy: A Benchmark for Explicit Physical Property Learning in Multi-Object Trajectory Forecastinghttps://blog.schoergendorfer.com/post/optimal-skill-selection-for-llm-agents-with-provable-bicriteria-guaranteesSecFeedde2026-08-21T04:16:30.035731+00:00Optimal Skill Selection for LLM Agents with Provable Bicriteria Guaranteeshttps://blog.schoergendorfer.com/post/rethinking-patch-based-multivariate-time-series-forecasting-with-semantic-structSecFeedde2026-08-21T04:16:29.993653+00:00Rethinking Patch Based Multivariate Time Series Forecasting with Semantic Structured Partitioninghttps://blog.schoergendorfer.com/post/learning-early-to-final-solution-consistency-for-milp-accelerationSecFeedde2026-08-21T04:16:29.975603+00:00Learning Early-to-Final Solution Consistency for MILP Accelerationhttps://blog.schoergendorfer.com/post/a-strong-linear-baseline-for-whole-heart-cardiac-shape-completion-on-ct-with-anSecFeedde2026-08-21T04:16:29.956693+00:00A Strong Linear Baseline for Whole-Heart Cardiac Shape Completion on CT, with an Open Eleven-Structure Statistical Shape Modelhttps://blog.schoergendorfer.com/post/eximo-vlm-guided-exploration-of-vla-policiesSecFeedde2026-08-21T04:16:29.902652+00:00EXIMO: VLM Guided Exploration of VLA Policieshttps://blog.schoergendorfer.com/post/write-once-run-everywhere-the-axon-dsl-for-shape-safe-and-framework-agnostic-llmSecFeedde2026-08-21T04:16:29.883375+00:00Write Once, Run Everywhere: The Axon DSL for Shape-Safe and Framework-Agnostic LLM Architectureshttps://blog.schoergendorfer.com/post/testnav-pareto-guided-search-for-compositional-robustness-testingSecFeedde2026-08-21T04:16:29.865046+00:00TESTNAV: Pareto-Guided Search for Compositional Robustness Testinghttps://blog.schoergendorfer.com/post/envharness-awakening-static-worlds-for-agent-learningSecFeedde2026-08-21T04:16:29.843521+00:00EnvHarness: Awakening Static Worlds for Agent Learninghttps://blog.schoergendorfer.com/post/policyguide-from-guarding-one-action-to-guiding-the-whole-workflow-for-policy-coSecFeedde2026-08-21T04:16:29.824699+00:00PolicyGuide: From Guarding One Action to Guiding the Whole Workflow for Policy-Compliant LLM Agentshttps://blog.schoergendorfer.com/post/sapo-single-rollout-autoregressive-policy-optimization-for-agentic-reinforcementSecFeedde2026-08-21T04:16:29.803801+00:00SAPO: Single-Rollout Autoregressive Policy Optimization for Agentic Reinforcement Learninghttps://blog.schoergendorfer.com/post/specification-delta-driven-data-governance-an-empirical-study-of-the-guillemotleSecFeedde2026-08-21T04:16:29.783627+00:00Specification-delta-driven data governance: an empirical study of the {\guillemotleft}spec-delta{\guillemotright} as the unit of change in lakehouse data platformshttps://blog.schoergendorfer.com/post/causal-reasoning-with-bipartite-graphical-causal-modelsSecFeedde2026-08-21T04:16:29.740351+00:00Causal Reasoning with Bipartite Graphical Causal Modelshttps://blog.schoergendorfer.com/post/adapt-physics-aware-diffusion-based-world-models-for-adaptive-predictive-transfeSecFeedde2026-08-21T04:16:29.714917+00:00ADAPT: Physics-Aware Diffusion-based World Models for Adaptive Predictive Transferable HVAC Controlhttps://blog.schoergendorfer.com/post/towards-general-embodied-intelligence-integrating-large-language-models-knowledgSecFeedde2026-08-21T04:16:29.691313+00:00Towards general embodied intelligence: integrating large language models, knowledge bases, and reasoning capabilities to build the next generation of AI agentshttps://blog.schoergendorfer.com/post/llms-as-acquisition-policies-for-finite-pool-materials-optimization-a-controlledSecFeedde2026-08-21T04:16:29.666309+00:00LLMs as Acquisition Policies for Finite-Pool Materials Optimization: A Controlled Studyhttps://blog.schoergendorfer.com/post/tt-net-quantum-inspired-tensor-network-denoising-in-conditional-gansSecFeedde2026-08-21T04:01:59.394105+00:00TT-net: Quantum Inspired Tensor Network Denoising in Conditional GANshttps://blog.schoergendorfer.com/post/genmatch-an-end-to-end-generative-matching-framework-for-micro-view-order-dispatSecFeedde2026-08-21T04:01:59.371537+00:00GenMatch: An End-to-End Generative Matching Framework for Micro-View Order-Dispatching in Ride-Hailinghttps://blog.schoergendorfer.com/post/safebranch-branch-pair-safety-alignment-for-embodied-agentsSecFeedde2026-08-21T04:01:59.341244+00:00SafeBranch: Branch-Pair Safety Alignment for Embodied Agentshttps://blog.schoergendorfer.com/post/beyond-memory-majority-latent-source-reasoning-for-multi-agent-memory-arbitratioSecFeedde2026-08-21T04:01:59.317617+00:00Beyond Memory Majority: Latent-Source Reasoning for Multi-Agent Memory Arbitrationhttps://blog.schoergendorfer.com/post/rethinking-the-evaluation-and-optimization-of-llm-based-social-simulationSecFeedde2026-08-21T04:01:59.292994+00:00Rethinking the Evaluation and Optimization of LLM-Based Social Simulationhttps://blog.schoergendorfer.com/post/learning-hierarchical-skill-policies-with-offline-quality-diversity-reinforcemenSecFeedde2026-08-21T04:01:59.267472+00:00Learning Hierarchical Skill Policies with Offline Quality-Diversity Reinforcement Learninghttps://blog.schoergendorfer.com/post/frequency-aware-continual-learning-for-smart-contract-vulnerability-detection-wiSecFeedde2026-08-21T04:01:59.245209+00:00Frequency-Aware Continual Learning for Smart Contract Vulnerability Detection with Large Language Modelshttps://blog.schoergendorfer.com/post/can-agent-memory-systems-track-evolving-stateSecFeedde2026-08-21T04:01:59.223193+00:00Can Agent Memory Systems Track Evolving State?https://blog.schoergendorfer.com/post/scientific-data-skills-enabling-agent-ready-scientific-data-services-at-scaleSecFeedde2026-08-21T04:01:59.200747+00:00Scientific Data Skills: Enabling Agent-Ready Scientific Data Services at Scalehttps://blog.schoergendorfer.com/post/enforcing-llm-safety-through-dmd-based-classification-of-prompt-response-embeddiSecFeedde2026-08-21T04:01:59.179595+00:00Enforcing LLM Safety through DMD-based Classification of Prompt-Response Embedding Dynamicshttps://blog.schoergendorfer.com/post/from-retrieved-context-to-runtime-control-adaptive-compression-for-edge-based-raSecFeedde2026-08-21T04:01:59.158537+00:00From Retrieved Context to Runtime Control: Adaptive Compression for Edge-based RAGhttps://blog.schoergendorfer.com/post/symposium-trust-via-auditable-records-for-communities-of-ai-scientist-agentsSecFeedde2026-08-21T04:01:59.133659+00:00Symposium: Trust via Auditable Records for Communities of AI Scientist Agentshttps://blog.schoergendorfer.com/post/beyond-imitation-filtering-on-policy-distillation-by-reasoning-progressSecFeedde2026-08-21T04:01:59.108018+00:00Beyond Imitation: Filtering On-Policy Distillation by Reasoning Progresshttps://blog.schoergendorfer.com/post/outcome-monitors-recovery-affordances-for-silent-tool-failuresSecFeedde2026-08-21T04:01:59.084383+00:00Outcome Monitors: Recovery Affordances for Silent Tool Failureshttps://blog.schoergendorfer.com/post/air-traffic-control-using-large-language-models-prompt-engineering-architectureSecFeedde2026-08-21T04:01:59.059326+00:00Air Traffic Control Using Large Language Models: Prompt Engineering, Architecture, and Evaluationhttps://blog.schoergendorfer.com/post/bounded-sovereignty-and-the-control-tax-pricing-ai-oversight-when-the-deployer-dSecFeedde2026-08-21T04:01:59.035010+00:00Bounded Sovereignty and the Control Tax: Pricing AI Oversight When the Deployer Does Not Own the Modelhttps://blog.schoergendorfer.com/post/interaction-valence-reveals-contrasting-social-networks-in-dairy-cattleSecFeedde2026-08-21T04:01:59.010210+00:00Interaction valence reveals contrasting social networks in dairy cattlehttps://blog.schoergendorfer.com/post/robust-metaheuristics-under-uncertainty-for-berth-allocation-and-quay-crane-assiSecFeedde2026-08-21T04:01:58.968774+00:00Robust Metaheuristics under Uncertainty for Berth Allocation and Quay Crane Assignment: A Reviewhttps://blog.schoergendorfer.com/post/active-inference-as-context-acquisition-for-ai-agentsSecFeedde2026-08-21T04:01:58.945671+00:00Active Inference as Context Acquisition for AI Agentshttps://blog.schoergendorfer.com/post/aws-security-agent-now-part-of-aws-continuum-now-supports-budget-controls-and-fiSecFeedde2026-08-21T03:16:25.430322+00:00AWS Security Agent (now part of AWS Continuum) now supports budget controls and finding revalidationhttps://blog.schoergendorfer.com/post/who-got-missed-in-the-mfa-rolloutx3f-more-powershell-x2b-graph-x2b-entra-scriptiSecFeedde2026-08-21T02:31:49.056387+00:00Who Got Missed in the MFA Rollout? More Powershell + Graph + Entra scripting!, (Fri, Aug 21st)https://blog.schoergendorfer.com/post/even-moar-powershell-looking-at-entra-logins-the-good-the-bad-and-the-password-sSecFeedde2026-08-21T02:01:28.646316+00:00Even MOAR Powershell, looking at Entra logins - the good, the bad and the password sprays, (Fri, Aug 21st)https://blog.schoergendorfer.com/post/chatgpt-search-now-uses-the-siteoperator-at-scaleSecFeedde2026-08-21T00:16:36.335729+00:00ChatGPT search now uses the site:operator at scalehttps://blog.schoergendorfer.com/post/usn-8666-1-linux-kernel-vulnerabilitiesSecFeedde2026-08-21T00:02:31.483334+00:00USN-8666-1: Linux kernel vulnerabilitieshttps://blog.schoergendorfer.com/post/usn-8663-1-linux-kernel-nvidia-vulnerabilitiesSecFeedde2026-08-20T23:46:30.817582+00:00USN-8663-1: Linux kernel (NVIDIA) vulnerabilitieshttps://blog.schoergendorfer.com/post/usn-8664-1-linux-kernel-nvidia-baseos-vulnerabilitiesSecFeedde2026-08-20T23:46:30.772742+00:00USN-8664-1: Linux kernel (NVIDIA BaseOS) vulnerabilitieshttps://blog.schoergendorfer.com/post/usn-8665-1-linux-kernel-raspberry-pi-vulnerabilitiesSecFeedde2026-08-20T23:46:30.726861+00:00USN-8665-1: Linux kernel (Raspberry Pi) vulnerabilitieshttps://blog.schoergendorfer.com/post/a-shot-scraper-style-json-api-on-bun-14s-new-bunwebviewSecFeedde2026-08-20T23:32:29.544015+00:00A shot-scraper-style JSON API on Bun 1.4's new Bun.WebViewhttps://blog.schoergendorfer.com/post/usn-8543-2-wget-regressionSecFeedde2026-08-20T23:32:29.518314+00:00USN-8543-2: Wget regression