Research
Great paper from Google and colleagues. It proposes an interesting approach to improve agents on long-horizon tasks. Long-running agents slo…
Great paper from Google and colleagues. It proposes an interesting approach to improve agents on long-horizon tasks. Long-running agents slow down and start poisoning their own context. Both symptoms
Great paper from Google and colleagues. It proposes an interesting approach to improve agents on long-horizon tasks. Long-running agents slow down and start poisoning their own context. Both symptoms come from the same design choice, which is keeping execution alive by appending every observation, action, and reasoning trace to a growing conversation. SKILL.state replaces that history with an explicit mutable execution state. At each step the model sees only the immutable skill specification, the current structured state, and the latest observation. Intermediate reasoning gets discarded the moment it produces a validated state update, so the prompt stops growing with the run. Across several datasets, models, and execution environments, task accuracy improves while cumulative token consumption drops. The abstraction is architecture-agnostic, so it ports into existing skill runtimes. Paper: https://arxiv.org/abs/2608.26263 Chat with Paper: https://academy.dair.ai/papers/explicit-execution-state-replaces-append-only-history-2608.26263
Related
- You don't often see one-word titles in AI papers. That aside, strong recommend this paper from Google DeepMind. I think this is an interesti…
- On Training Large Language Models for Long-Horizon Tasks: An Empirical Study of Horizon Length
- MineEvolve: Self-Evolution with Accumulated Knowledge for Long-Horizon Embodied Minecraft Agents
- HeteroCache: A Dynamic Retrieval Approach to Heterogeneous KV Cache Compression for Long-Context LLM Inference
Source: DAIR.AI (X) | 2026-08-31