Long conversational agents improve memory with energy-based schema updates
MINDSET: Energy-based Schema Evolution for Long Conversational Agent Memory
Artificial Intelligence
Summary
Conversational agents, like digital assistants, need to remember past conversations to help users effectively. The authors show that these agents can manage memory better by storing conversations as unchangeable episodes and grouping them into versions called schemas. Their method, called MINDSET, decides how to update these schemas based on how much new information conflicts or changes old info. This approach helps the agent keep both old and new knowledge clearly, improving its answers without reprocessing everything all the time.
What this means in practice
- •For conversational ai developers: Build chatbots that adapt their memory to changes in user instructions over long conversations, improving response relevance and reducing repeated context requests.
- •For customer support platform teams: Enhance automated support agents to better recall and update long-term interactions, aiding in faster and more accurate issue resolution.
Authors
Sujato Dutta, Sreekruthy Tummala, Shashank Vanga, Ayushmi Pavani
Abstract
Long conversational agents have become essential in our daily lives. They must remember what was said long back in order to help us efficiently complete a task without needing the user to repeat instructions and context repeatedly. However, the main issue is that instructions and context change over time and so the agents must be able to adapt accordingly. A useful memory system should preserve both current and historical states, distinguish stale information from active knowledge, retrieve evidence appropriate to the query and avoid repeatedly invoking a large language model to rewrite prior interactions. We introduce MINDSET, a memory controller that stores a conversation as immutable episodes and organizes them into versioned schemas through minimum-energy state transitions. Each incoming episode may reinforce, supersede, split or create a schema. The transition decision balances representation distortion, contradiction, historical damage, fragmentation and internal inconsistency, while hysteresis prevents isolated contradictions from prematurely rewriting stable memory. We evaluate MINDSET against 5 memory systems on a reproducible sample of 850 questions (700 LoCoMo + 150 MemoryAgentBench). MINDSET obtains the highest observed LoCoMo answer F1 while significantly improving retrieval ranking (Recall@8, MRR and nDCG@8) over the second best method LightMem (p<0.01 after Holm correction). It obtains the highest observed scores on MemoryAgentBench although the relative difference is low. Ablations identify controlled fragmentation and schema-aware assignment as the largest contributors to answer quality. Additionally, a 700-question cross-model evaluation with GLM-4.7 and Gemma-4-31B supported model independence. These results show that long-term memory can be better handled as constrained state management rather than continual summarization.