Direct Relational Set-Risk Pruning is introduced, which formulates agent-history compression as risk-constrained selection over deletion sets and shows that decision-conditioned relations, retained-context information, pair interactions, and abstention each contribute to reliable pruning.
Abstract
Long-horizon language-model agents accumulate reasoning traces, tool exchanges, and observations whose relevance changes with the current decision. Existing compression strategies often score historical units independently, but the safety of deleting several units is generally not determined by their singleton scores: redundant evidence, accumulated small effects, and the information that remains after deletion all matter. We introduce Direct Relational Set-Risk Pruning (DRSR), which formulates agent-history compression as risk-constrained selection over deletion sets. Offline, DRSR constructs exact counterfactual supervision by jointly deleting protocol-valid history Blocks and measuring the change in teacher-forced likelihood of the same recorded next output. A lightweight scorer then predicts set-level harm from online-visible relations between candidate history and the current pre-action state, together with deleted-retained and pairwise set structure. At deployment, DRSR evaluates a small set of structurally valid deletion candidates with the lightweight scorer and removes the largest feasible set under recency, protocol, budget, and learned-risk constraints, abstaining when no set is sufficiently safe. On WorkBuddyBench Full260, DRSR increases mean reward from 0.699 to 0.802 while reducing total model tokens by 20.820%. On the fixed Eval40 comparison, it obtains 0.794 reward at 1.211M tokens per task, using 35.850% fewer tokens than the uncompressed agent. Mechanistic analyses and ablations further show that decision-conditioned relations, retained-context information, pair interactions, and abstention each contribute to reliable pruning.
A novel and scalable framework that rewrites a coding agent's context as the repository changes rather than as the context grows, and can achieve a higher resolve rate in all experiments spanned by six coding agents and three edit-heavy benchmarks with little computational overhead.
Experiments establish ReTree as an effective self-correcting memory abstraction for long-horizon search, and show that ReTree consistently outperforms Full-Trajectory ReAct in question-answering and search benchmarks.
Aijun Yang, Qianxue Guo, Ziyi Huang et al.· 0 citations
Results characterize agent reasoning as dynamic working state rather than permanent interaction history, and suggest that historical reasoning becomes more replaceable once task relevant derived state has been reliably externalized into code, files, tool outputs, or environmental feedback.
Ming-Xuan Wang, Fei Luo, Bo Wang et al.· arXiv.org· 4 citations
Retrieval assembles repository context by ranking passages for relevance to the current query. A coding agent halfway through an issue has already read much of what such a ranker returns. Relevance is scored per passage, but sufficiency belongs to the set: independently scored passages can fill the budget with support...
Zhe-Xi Feng, Rui-Yi Zhang, Yong-Bo Yang et al.· arXiv.org· 0 citations
Experiments on WebShop, ALFWorld, and visual Sokoban show consistent improvements across language and vision-language models, while diagnostic ablations support the effectiveness of Bellman fixed-point value estimation and show that step-level credit should be incorporated selectively rather than uniformly into the fin...
Hongxi Yan, Ziyue Huang, Shichao Fan et al.· 1 citation
This work proposes State Conditioned Compression (StateComp), a framework that determines when historical interactions can be safely compressed according to the current agent state, and demonstrates that it reduces total agent and summarization tokens while maintaining task performance, and achieves a 12.67-fold speedu...
By studying the computational foundations of language learning, Associate Professor Jacob Andreas builds intelligent systems that can communicate effectively with humans.
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.