Artificial Intelligence · 25.08.2026, 05:01 UTC
CausalCache: Conditional High-Fidelity Restoration for Long-Horizon GUI Agents
| Schweregrad | info |
|---|---|
| Kategorie | Artificial Intelligence |
| Quelle | arXiv cs.AI ↗ |
| Veröffentlicht | 25.08.2026 UTC |
Sicherheitsmeldung mit Schweregrad noch nicht bewertet. Technische Details im Tab „Originaltext“; empfohlene Schritte in der Checkliste.
arXiv:2608.22577v1 Announce Type: new Abstract: Long-horizon GUI agents can retain a complete interaction trace cheaply as textual action records, but expose only a few past events to the policy in high-fidelity pixels. We formulate this as conditional fidelity restoration: each event persists in summary-only form and is linked to an archived screenshot, while an active visual-context budget $B$ limits how many events may be promoted to summary-plus-image form. Recent-$B$ spends every slot on the latest events. CausalCache instead reallocates the same $B$ promotions over the complete trace, evicting a recent image only when a distant event has higher conditional marginal utility. Its history-gated key/value (HGKV) adapter modifies only restored history-image tokens and is exactly bypassed with no history image. Matched-budget replacement groups and per-arm-anchored difference-in-differences supervision make uniform history amplification worth zero; a budget-aware selector then chooses which summarized events to restore. On desktop, the frozen policy shows no reliable preference for a task-relevant archived screenshot over the recent frame it would displace; HGKV learns exactly that selectivity inside a pre-specified drift envelope. On OSWorld-Verified, restoring history to high fidelity is worth about $13$ success points over summary-only memory, while same-budget allocations remain indistinguishable. Zero-shot on a cross-application mobile benchmark, CausalCache significantly improves overall success over the same-budget recent allocation ($+3.7$ points on the full …
Maßnahmen
⬇ Als MarkdownVerwandte Beiträge
- info Deep Learning-Based Multi-User Communication Design for Dense IoT Networks: Interference-Aware Finite-Blocklength Communication and Preliminary MIMO Extensions
- info Safety Hacking in Constrained Best-of-$N$ Inference-time Scaling
- info AraDetox: A Multi-Dialect Arabic Detoxification Dataset
- info Fairness-Aware Mixture-of-Experts via Subgroup Reweighting and Gate Regularization