paper-with-me

홈 › Papers

Learning What to Remember: Observability-Safe Memory Retention via Constrained Optimization for Long-Horizon Language Agents

2026-06-09 · Qingcan Kang, Liu Mingyang, Shixiong Kai, Kaichao Liang, Tao Zhong, Mingxuan Yuan arxiv

Long-horizon language agents accumulate observations, reasoning traces, and retrieved facts exceeding context windows, making memory retention a fundamental resource-allocation problem. Existing systems treat retention as local and do not model long-term consequences under observability constraints. To fill this gap, we formulate memory retention as a constrained stochastic optimization with budget feasibility, evidence utility, and delayed costs including miss, reacquisition, and stale penalties. We show this multi-step problem is NP-hard, making exact solution intractable. Moreover, deployment decisions must be made under partial observability. To address these challenges, we propose OSL-MR (Observability-Safe Learning for Memory Retention), a learning-augmented framework that enforces a strict separation between online-observable features and offline-available supervision. OSL-MR combines an evidence learner trained from realized evidence with a Mixed-Score heuristic that serves as a deployable online-safe baseline and an inductive prior. The policy learns query-conditioned evidence from interaction data and remains deployable under the same constraints. Experiments on LoCoMo and LongMemEval show OSL-MR outperforms recency-based, Generative Agents-style, and other heuristic baselines, especially under tight budgets. The Mixed-Score prior improves precision and recall, and sensitivity analysis shows robustness across cost settings. On small solvable instances, single-step optimization is insufficient to anticipate future demand shifts, while OSL-MR stays significantly closer to the dynamic-programming optimum, confirming the necessity of the sequential formulation and reinforcing our learning-guided approximation. These results establish constrained stochastic optimization and optimization-guided learning as a principled foundation for memory management in long-horizon agents.

📄 PDF Abstract BibTeX arXiv:2606.10616

Code (0)

등록된 구현이 없습니다.

Tasks

Stochastic Optimization

Similar Papers 제목 키워드 기반

Learning What to Remember: Adaptive Probabilistic Memory Retention for Memory-Efficient Language Models

2025-10-09 · S M Rafiuddin, Muntaha Nujat Khan arxiv

Transformer attention scales quadratically with sequence length O(n^2), limiting long-context use. We propose Adaptive Retention, a probabilistic, layer-wise token selection mechanism that learns which representations to…

Document Summarization

Learning What to Remember: Long-term Episodic Memory Networks for Learning from Streaming Data

2018-12-11 · ICLR 2019 5 · Hyunwoo Jung, Moonsu Han, Minki Kang, Sungju Hwang

Current generation of memory-augmented neural networks has limited scalability as they cannot efficiently process data that are too large to fit in the external memory storage. One example of this is lifelong learning sc…

Lifelong learningQuestion AnsweringScheduling

The act of remembering: a study in partially observable reinforcement learning

2020-10-05 · Rodrigo Toro Icarte, Richard Valenzano, Toryn Q. Klassen, Phillip Christoffersen 외

Reinforcement Learning (RL) agents typically learn memoryless policies---policies that only consider the last observation when selecting actions. Learning memoryless policies is efficient and optimal in fully observable …

Partially Observable Reinforcement Learningreinforcement-learningReinforcement Learning (RL)

Recognition, recall, and retention of few-shot memories in large language models

2023-03-30 · A. Emin Orhan

The training of modern large language models (LLMs) takes place in a regime where most training examples are seen only a few times by the model during the course of training. What does a model remember about such example…

Learning What to Remember: A Cognitively Grounded Multi-Factor Value Model for Agentic Memory

2026-06-11 · Zhibao Chen, Qian Cheng arxiv

Long-running LLM agents accumulate interaction histories far larger than any context window, forcing a standing decision: what to encode deeply, what to forget, and what to retrieve under a fixed memory budget. Productio…

Semantic Similarity