paper-with-me

Papers

Decision-Aware Memory Cards: Counterfactual-Inspired Context Selection and Compression for Tool-Using LLM Agents

2026-06-06 · Xinyu Guan, Qianyang Zhao, Yuming Deng arxiv

Modern large language model (LLM) agents do not simply need longer contexts; they need decision-relevant evidence at the moment of action. We study decision-aware context selection: ranking retrieved files, tests, traces, rules, and memories by their expected effect on an agent's next action rather than by semantic similarity alone. We present the Counterfactual-Inspired Context Layer (CICL), which builds an instance context graph, estimates decision-oriented utility for candidate units, and compresses selected evidence into typed memory cards. The same schema can be instantiated with hosted LLM judges, local surrogates, or lightweight rankers, making the selection protocol auditable across model choices. On 50 SWE-bench Verified file-retrieval instances, Qwen3.6-Plus reranking of BM25 top-50 candidates improves hit@1 from 0.58 to 0.78 and MRR@10 from 0.634 to 0.790, with all 2,500 judgments parseable. Controlled diagnostics show that CICL identifies action-critical evidence: removing the top-utility semantic unit reduces F1 from 0.245 to 0.000. In selected-then-compressed mode, memory cards save 44.93 tokens per query while preserving selected evidence. CICL provides a practical layer for measuring, ranking, and compressing decision-critical context for tool-using agents. Code is available at https://github.com/stephen-guan-researcher/CICL.

📄 PDF Abstract BibTeX arXiv:2606.08151

Code (0)

등록된 구현이 없습니다.

Tasks

Semantic Similarity

Similar Papers 제목 키워드 기반

We Need Explanation Cards to Connect Explanation Algorithms to the Real World

2026-06-15 · Eric Günther, Balázs Szabados, Kristof Meding, Gunnar König 외 arxiv

Algorithmic explanations are intended to help stakeholders understand opaque algorithmic decisions, but in practice, they often fall short. First, the meaning of algorithmic explanations is often not what one might intui…

FinCARDS: Card-Based Analyst Reranking for Financial Document Question Answering

2026-01-11 · Yixi Zhou, Fan Zhang, Yu Chen, Haipeng Zhang 외 arxiv

Financial question answering (QA) over long corporate filings requires evidence to satisfy strict constraints on entities, financial metrics, fiscal periods, and numeric values. However, existing LLM-based rerankers prim…

Question Answering

Latent State Design for World Models under Sufficiency Constraints

2026-05-03 · Keon Woo Kim arxiv

A world model matters to an agent only through the state it constructs. That state must preserve some information, discard other information, and support some future function: prediction, control, planning, memory, groun…

Video Prediction

OCTET: Object-aware Counterfactual Explanations

2022-11-22 · CVPR 2023 1 · Mehdi Zemni, Mickaël Chen, Éloi Zablocki, Hédi Ben-Younes 외

Nowadays, deep vision models are being widely deployed in safety-critical applications, e.g., autonomous driving, and explainability of such models is becoming a pressing concern. Among explanation methods, counterfactua…

Autonomous DrivingcounterfactualCounterfactual ExplanationExplanation Generation+2

Learning to Control Coupled-Dynamics Environments with Joint Markov Decision Processes

2026-08-24 · Ege C. Kaya, Aliasghar Pourghani, Mahsa Ghasemi, Vijay Gupta 외 arxiv

Coupled-dynamics environments expose the one-step outcomes that would follow from several possible counterfactual actions under a common realization of exogenous randomness. The ordinary Markov decision process formalism…