Remember the Decision, Not the Description: A Rate-Distortion Framework for Agent Memory
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909033557917696 |
|---|---|
| author | Zou, Mingxi Guo, Zhihan Liang, Langzhang Wang, Zhuo Wang, Qifan Wen, Qingsong King, Irwin Qu, Lizhen Xu, Zenglin |
| author_facet | Zou, Mingxi Guo, Zhihan Liang, Langzhang Wang, Zhuo Wang, Qifan Wen, Qingsong King, Irwin Qu, Lizhen Xu, Zenglin |
| contents | Long-horizon language agents must operate under limited runtime memory, yet existing memory mechanisms often organize experience around descriptive criteria such as relevance, salience, or summary quality. For an agent, however, memory is valuable not because it faithfully describes the past, but because it preserves the distinctions between histories that must remain separated under a fixed budget to support good decisions. We cast this as a decision-centric rate-distortion problem, measuring memory quality by the loss in achievable decision quality induced by compression. This yields an exact forgetting boundary for what can be safely forgotten, and a memory-distortion frontier characterizing the optimal tradeoff between memory budget and decision quality. Motivated by this decision-centric view of memory, we propose DeMem, an online memory learner that refines its partition only when data certify that a shared state would induce decision conflict, and prove near-minimax regret guarantees. On both controlled synthetic diagnostics and long-horizon conversational benchmarks, DeMem yields consistent gains under the same runtime budget, supporting the principle that memory should preserve the distinctions that matter for decisions, not descriptions. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_10870 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Remember the Decision, Not the Description: A Rate-Distortion Framework for Agent Memory Zou, Mingxi Guo, Zhihan Liang, Langzhang Wang, Zhuo Wang, Qifan Wen, Qingsong King, Irwin Qu, Lizhen Xu, Zenglin Artificial Intelligence Long-horizon language agents must operate under limited runtime memory, yet existing memory mechanisms often organize experience around descriptive criteria such as relevance, salience, or summary quality. For an agent, however, memory is valuable not because it faithfully describes the past, but because it preserves the distinctions between histories that must remain separated under a fixed budget to support good decisions. We cast this as a decision-centric rate-distortion problem, measuring memory quality by the loss in achievable decision quality induced by compression. This yields an exact forgetting boundary for what can be safely forgotten, and a memory-distortion frontier characterizing the optimal tradeoff between memory budget and decision quality. Motivated by this decision-centric view of memory, we propose DeMem, an online memory learner that refines its partition only when data certify that a shared state would induce decision conflict, and prove near-minimax regret guarantees. On both controlled synthetic diagnostics and long-horizon conversational benchmarks, DeMem yields consistent gains under the same runtime budget, supporting the principle that memory should preserve the distinctions that matter for decisions, not descriptions. |
| title | Remember the Decision, Not the Description: A Rate-Distortion Framework for Agent Memory |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2605.10870 |