Tencent open-sources T-Mem, a memory architecture that anticipates instead of archiving
blaizedsouza · x · 2026-09-21
Tencent released T-Mem (EMNLP paper, MIT license), claiming it's the first open-source conversational memory architecture to cover associative recall, not just similarity search. Existing systems fail without keyword overlap: mainstream systems drop 28-50 points on LoCoMo-Plus, while T-Mem drops only 5.45.
- Write time: each memory ships with a trigger predicting when it will matter again
- Read time: triggers pull in memories by situation, even with zero keyword overlap
- Architecture layer: plugs into any LLM without fine-tuning
Related event: Tencent Open-Sources T-Mem, an Associative-Memory Architecture for LLMs(4 posts)→
More from coding & agent
- User automates entire video clipping workflow with Nous Research's Hermes agent — Teknium · 2026-11-12
- As models improve, writing good evals gets harder and more important — amankhan · 2026-11-12
- Prime agent ships with day-one multi-agent communication, runs in Minecraft unmodified — sarahcat21 · 2026-11-12
- Autonomous agent edits just 1,024 weights to lift Qwen3.5 0.8B from 7.4% to 37.4% on MATH-500 — VictorSanh · 2026-11-12
- AI game jam with Anthropic, Google, OpenAI teams: cash prizes plus $50K continuation grants — chrisfirst · 2026-11-12
- Free AI game jam in SF Nov 14–15 with teams from Anthropic, Google, Meshy, OpenAI — chrisfirst · 2026-11-12