跳到正文
arXiv cs.AI· Guolong Zheng, Chen Hou, Xu Yang, Jianshan Zhang, Tao Huang·· 10 小时前AI 评分53

Semantic Context Handoff:面向长程智能体的按需证据上下文交接

Semantic Context Handoff with On-Demand Evidence for Long-Horizon Agents

AI 导读

论文提出 Semantic Context Handoff(SCH),核心是把智能体的语义状态与支撑动作-观察证据分离,证据可按需重新进入上下文,无需单独摘要或检索模型调用。

正文 · 原文

View PDF HTML (experimental)

Abstract:Traditional long-horizon LLM agents often hand off context through full-history prompting, retaining all previous action--observation interactions. As execution proceeds, this history keeps growing, increasing inference cost and potentially degrading accuracy. Existing methods control this growth by preserving compact conclusions while removing earlier observations, but doing so can discard evidence that later becomes relevant. We introduce Semantic Context Handoff (SCH), whose central abstraction is state--evidence separation: a compact semantic state carries the agent's current interpretation, while supporting action--observation evidence remains separately addressable and can re-enter context when later execution makes it relevant, without separate summarization or retrieval-model calls.
Empirically, SCH improves long-horizon agent performance while substantially reducing context cost. In a controlled 150-task SWE-bench study, SCH's evidence recovery improves Pass@1 from $80.0\%$ to $86.9\%$ at essentially unchanged token use and execution time. In the complete-system comparison, SCH achieves 86.9% Pass@1 while using 43% fewer tokens and 41% less wall-clock time than the strongest baseline by task performance. Across models and benchmarks, SCH consistently reduces token use relative to full-history prompting while remaining competitive with or improving upon the evaluated baselines. Trajectory analysis confirms that SCH recovers evidence from substantially earlier parts of the interaction history rather than merely preserving a recent context window.
Subjects: Artificial Intelligence (cs.AI)
Cite as: arXiv:2611.24357 [cs.AI]
  (or arXiv:2611.24357v1 [cs.AI] for this version)
  https://doi.org/10.48550/arXiv.2611.24357

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Tao Huang [view email]
[v1] Wed, 18 Nov 2026 12:49:41 UTC (1,358 KB)

来源:arXiv cs.AI · arxiv.org