mlsanityData to 7 Oct 2026

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

arXiv 2606.08151 ↗First posted 6 Jun 2026cs.AIcited by 1

On the map: LLM agents and tool use → Agent memory systems

Abstract (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.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Placed in Agent memory systems, one of 16 directions in LLM agents and tool use. Builds on 38 of its 43 references that are in the map. First posted on arXiv 6 Jun 2026.

Builds on

38 of 43 references.

Closest ancestors · 6

Same direction, newest first

  1. 2026EvoMemBench: Benchmarking Agent Memory from a Self-Evolving Perspective18 May
  2. 2026Causal Intervention-Based Memory Selection for Long-Horizon LLM Agents17 May
  3. 2026MemoryArena: Benchmarking Agent Memory in Interdependent Multi-Session Agentic Tasks18 Feb
  4. 2025Agentic Context Engineering: Evolving Contexts for Self-Improving Language Models6 Oct
  5. 2025ACON: Optimizing Context Compression for Long-horizon LLM Agents1 Oct
  6. 2025Evaluating Memory in LLM Agents via Incremental Multi-Turn Interactions7 Jul

Other references · 32

  1. 2022Training language models to follow instructions with human feedbackcited by 15,712
  2. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  3. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
  4. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,355
  5. 2023Toolformer: Language Models Can Teach Themselves to Use Toolscited by 4,072
  6. 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,064
  7. 2020Dense Passage Retrieval for Open-Domain Question Answeringcited by 3,779
  8. 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,383
  9. 2023SWE-bench: Can Language Models Resolve Real-World GitHub Issues?cited by 3,193
  10. 2020Learning to summarize from human feedbackcited by 2,562
  11. 2017Billion-scale similarity search with GPUscited by 2,477
  12. 2023Voyager: An Open-Ended Embodied Agent with Large Language Modelscited by 2,081
20 more
  1. 2023AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversationcited by 1,906
  2. 2023Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflectioncited by 1,800
  3. 2024SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineeringcited by 1,628
  4. 2023LongBench: A Bilingual, Multitask Benchmark for Long Context Understandingcited by 1,378
  5. 2021Unsupervised Dense Information Retrieval with Contrastive Learningcited by 1,203
  6. 2020ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERTcited by 1,160
  7. 2024OpenHands: An Open Platform for AI Software Developers as Generalist Agentscited by 956
  8. 2022Atlas: Few-shot Learning with Retrieval Augmented Language Modelscited by 940
  9. 2023ChatDev: Communicative Agents for Software Developmentcited by 893
  10. 2019CodeSearchNet Challenge: Evaluating the State of Semantic Code Searchcited by 638
  11. 2022Precise Zero-Shot Dense Retrieval without Relevance Labelscited by 508
  12. 2023LongLLMLingua: Accelerating and Enhancing LLMs in Long Context Scenarios via Prompt Compressioncited by 443
  13. 2024Agentless: Demystifying LLM-based Software Engineering Agentscited by 400
  14. 2023RepoBench: Benchmarking Repository-Level Code Auto-Completion Systemscited by 346
  15. 2023LLMLingua: Compressing Prompts for Accelerated Inference of Large Language Modelscited by 279
  16. 2023Compressing Context to Enhance Inference Efficiency of Large Language Modelscited by 185
  17. 2026ContextBench: A Benchmark for Context Retrieval in Coding Agentscited by 23
  18. 2026SWE Context Bench: A Benchmark for Context Learning in Codingcited by 8
  19. 2026RepoShapley: Shapley-Enhanced Context Filtering for Repository-Level Code Completioncited by 3
  20. 2025AutoContext: Instance-Level Context Learning for LLM Agentscited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Plan Pointers and Record-Directive Form in Budgeted Verification of Inherited Agent Memorycited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026ContextWeaver: Selective and Dependency-Structured Memory Construction for LLM Agents
  2. 2026Weighted Memory Tree: Remembering What Matters for Long-Horizon LLM Agents
  3. 2026Memory for Autonomous LLM Agents:Mechanisms, Evaluation, and Emerging Frontiers
  4. 2026Akashic: A Low-Overhead LLM Inference Service with MemAttention
  5. 2026LLM Agents Are Latent Context Managers: Eliciting Self-Managed Context via State Proprioception
  6. 2026The Context Gathering Decision Process: A POMDP Framework for Agentic Search
  7. 2026Learning When to Remember: Risk-Sensitive Contextual Bandits for Abstention-Aware Memory Retrieval in LLM-Based Coding Agents
  8. 2026ElasticMem: Latent Memory as a Learnable Resource for LLM Agents