← 研踪 PaperTrailarXiv:2610.08586 · v2
MINDSET: Energy-based Memory Evolution for Long Conversational Agents
首次提交 2026-10-06 23:55
版本更新 2026-11-03 15:13
雷达观测 2026-11-04 23:40
摘要
In conversational memory, a later statement may replace an earlier fact, describe a different context, or be an isolated contradiction. Thus, agentic memory must track what changed over time without erasing what was true before. We introduce MINDSET, a memory controller that makes this distinction through explicit state transitions. It preserves original dialogue episodes and organizes them into versioned schemas. For each new episode, it chooses whether to reinforce a schema, supersede its current state, create a contextual branch, or start a new schema. A local energy score balances coherence, contradiction, historical preservation and fragmentation. Stability rules prevent isolated conflicts from immediately replacing established memory. On a frozen 700-question LoCoMo subset, MINDSET achieves the highest observed answer F1 among six evaluated systems and significantly improves retrieval ranking over LightMem. Without retuning, it obtains 69.23% judge accuracy on all LongMemEval knowledge-update cases, compared with 64.10% for Vector RAG. Two additional answer model backbones using Gemma-4-31B and GLM-4.7 provide further qualified evidence of generalization. These findings support explicit memory-state management as a useful approach to changing conversational knowledge.
观测到的版本
相邻记录可能跳过中间版本；首次发现 v2 及以上也会标为版本更新。
v2 · 2026-11-03 15:13 · 最新观测版本
MINDSET: Energy-based Memory Evolution for Long Conversational Agents
In conversational memory, a later statement may replace an earlier fact, describe a different context, or be an isolated contradiction. Thus, agentic memory must track what changed over time without erasing what was true before. We introduce MINDSET, a memory controller that makes this distinction through explicit state transitions. It preserves original dialogue episodes and organizes them into versioned schemas. For each new episode, it chooses whether to reinforce a schema, supersede its current state, create a contextual branch, or start a new schema. A local energy score balances coherence, contradiction, historical preservation and fragmentation. Stability rules prevent isolated conflicts from immediately replacing established memory. On a frozen 700-question LoCoMo subset, MINDSET achieves the highest observed answer F1 among six evaluated systems and significantly improves retrieval ranking over LightMem. Without retuning, it obtains 69.23% judge accuracy on all LongMemEval knowledge-update cases, compared with 64.10% for Vector RAG. Two additional answer model backbones using Gemma-4-31B and GLM-4.7 provide further qualified evidence of generalization. These findings support explicit memory-state management as a useful approach to changing conversational knowledge.
查看此版本解读 →查看此版本原文 ↗