# Causal Memory Policy: Making Memory Utility Identifiable by Intervening on Retrieval

arXiv: https://arxiv.org/abs/2610.02070 · HTML: https://mlsanity.com/paper/2610.02070/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 37 of its 61 references that are in the map.
- First posted on arXiv 1 Oct 2026.

## Builds on (most cited first, 30 of 37)

- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering](https://mlsanity.com/paper/1809.09600/) (2018)
- [Generative Agents: Interactive Simulacra of Human Behavior](https://mlsanity.com/paper/2304.03442/) (2023)
- [MemGPT: Towards LLMs as Operating Systems](https://mlsanity.com/paper/2310.08560/) (2023)
- [A-MEM: Agentic Memory for LLM Agents](https://mlsanity.com/paper/2502.12110/) (2025)
- [MuSiQue: Multihop Questions via Single-hop Question Composition](https://mlsanity.com/paper/2108.00573/) (2021)
- [Evaluating Very Long-Term Conversational Memory of LLM Agents](https://mlsanity.com/paper/2402.17753/) (2024)
- [Counterfactual Fairness](https://mlsanity.com/paper/1703.06856/) (2017)
- [Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory](https://mlsanity.com/paper/2504.19413/) (2025)
- [LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory](https://mlsanity.com/paper/2410.10813/) (2024)
- [Data Shapley: Equitable Valuation of Data for Machine Learning](https://mlsanity.com/paper/1904.02868/) (2019)
- [MemoryBank: Enhancing Large Language Models with Long-Term Memory](https://mlsanity.com/paper/2305.10250/) (2023)
- [Evaluating Memory in LLM Agents via Incremental Multi-Turn Interactions](https://mlsanity.com/paper/2507.05257/) (2025)
- [Memorizing Transformers](https://mlsanity.com/paper/2203.08913/) (2022)
- [Augmenting Language Models with Long-Term Memory](https://mlsanity.com/paper/2306.07174/) (2023)
- [Unbiased Learning-to-Rank with Biased Feedback](https://mlsanity.com/paper/1608.04468/) (2016)
- [ContextCite: Attributing Model Generation to Context](https://mlsanity.com/paper/2409.00729/) (2024)
- [Appropriate Reliance on AI Advice: Conceptualization and the Effect of Explanations](https://mlsanity.com/paper/2302.02187/) (2023)
- [Degenerate Feedback Loops in Recommender Systems](https://mlsanity.com/paper/1902.10730/) (2019)
- [STALE: Can LLM Agents Know When Their Memories Are No Longer Valid?](https://mlsanity.com/paper/2605.06527/) (2026)
- [A Survey on Data Synthesis and Augmentation for Large Language Models](https://mlsanity.com/paper/2410.12896/) (2024)
- [Memory Sandbox: Transparent and Interactive Memory Management for Conversational Agents](https://mlsanity.com/paper/2308.01542/) (2023)
- [Model Internals-based Answer Attribution for Trustworthy Retrieval-Augmented Generation](https://mlsanity.com/paper/2406.13663/) (2024)
- [LongMemEval-V2: Evaluating Long-Term Agent Memory Toward Experienced Colleagues](https://mlsanity.com/paper/2605.12493/) (2026)
- [MemPO: Self-Memory Policy Optimization for Long-Horizon Agents](https://mlsanity.com/paper/2603.00680/) (2026)
- [SkillAudit: Ground-Truth-Free Skill Evolution via Paired Trajectory Auditing](https://mlsanity.com/paper/2606.14239/) (2026)
- [Graph Neural Network Causal Explanation via Neural Causal Models](https://mlsanity.com/paper/2407.09378/) (2024)
- [Forget to Improve: On-Device LLM-Agent Continual Learning via Budget-Curated Memory](https://mlsanity.com/paper/2606.25115/) (2026)
- [Source Attribution in Retrieval-Augmented Generation](https://mlsanity.com/paper/2507.04480/) (2025)
- [Causal Intervention-Based Memory Selection for Long-Horizon LLM Agents](https://mlsanity.com/paper/2605.17641/) (2026)

## Abstract (arXiv)

Memory-augmented large language models must decide which memories to retain, and recent systems do so by estimating each memory's effect on task performance. However, these estimates rely entirely on retrieved memories. When a memory is never retrieved, store-level interventions produce identical outcomes, leaving its utility unidentified. This is a retrieval-level positivity violation, invisible to diagnostics that examine only memory operations. We introduce Causal Memory Policy (CMP), a causal framework that restores identification by intervening on retrieval itself, reserving a fixed number of context slots for memories sampled with known propensities. CMP estimates memory utility by self-normalized inverse propensity weighting under a balanced assignment design. We prove the causal factorization of memory utility through retrieval, the unbiasedness and exact variance of the estimator, and the optimal decision rule under irreversible operations. Empirically, identification fails for 54% of required memories on LongMemEval and 67% on LoCoMo, and the failure persists in a deployed memory system. CMP improves discrimination between required and non-required memories from 0.54 to 0.66 AUC. Finally, we show that identified memory utility alone is insufficient for retention decisions: per-query utility reaches 0.78 AUC on the query for which it is estimated, yet no aggregation available to a retention policy predicts a memory's value on unseen queries. Code is available at: https://anonymous.4open.science/r/cmp-release-D0C3/.