T-Mem: Memory That Anticipates, Not Archives

2026-06-13

AI Summary

This work addresses the limitation of current large language models in long-term dialogue memory, which rely heavily on surface-level similarity and struggle to retrieve semantically related memories lacking explicit lexical overlap. To overcome this, the authors propose T-Mem, a novel architecture that, for the first time, integrates episodic future thinking from cognitive science into dialogue memory systems. During memory encoding, T-Mem simulates future retrieval contexts to generate two types of cues: descriptive triggers based on lexical and dense vector similarity, and associative triggers derived from latent semantic arcs. This dual-channel mechanism enables effective recall at both single-fact and full-dialogue granularities. Evaluated on the LoCoMo and LoCoMo-Plus benchmarks, T-Mem achieves state-of-the-art performance, significantly improving recall of semantically linked memories.

Abstract

Long-term memory is essential for conversational agents to remain coherent across extended dialogues, follow through on commitments made many sessions earlier, and adapt their behaviour to each user. Current LLM-backed long-term conversational memory, however, is reachability-bounded by the similarity between a query and stored content, both lexical and dense-vector. The approach is effective when query and memory share surface features such as wording or named entities (we call this descriptive). But it misses another, equally valuable class of cases, where query and memory do not share surface features and are tied only by a latent semantic arc (associative). On this regime prevailing long-term memory systems collectively fail. Covering this other half is what allows an assistant, for the first time, to actively draw on past dialogue as a semantic asset. On the memory side, this is the engineering counterpart of what cognitive science calls episodic future thinking: rehearsing past experience for the future contexts under which it will need to be found. We call these write-time rehearsals triggers. We propose T-Mem, the first long-term conversational memory architecture that covers both descriptive and associative recall. At each of two evidence granularities, single facts and full exchanges, T-Mem instantiates one descriptive trigger family and one associative trigger family, so that every memory remains reachable from both surface-similar and relevance-bound queries. As empirical validation, T-Mem reaches state-of-the-art on both LoCoMo and LoCoMo-Plus.

Problem

• long-term memory
• conversational agents
• associative recall
• descriptive recall
• semantic retrieval

Innovation

• T-Mem
• associative memory
• episodic future thinking
• conversational memory
• trigger-based retrieval