MemCodex organizes an agent’s experience into a hierarchy of executable programs covering summaries, relational knowledge, reusable skills and latent memory.
Self-programmed memory adapts retrieval depth to each agent query
MemCodex evolves executable memory layers and consults them from coarse summaries to source history, stopping when it finds sufficient evidence.
Top University
Xiaoqiang Wang · Bang Liu
Université de Montréal · Mila – Quebec AI Institute
Research Digest··2 min read
Wang and Liu present MemCodex, an agent memory architecture that can rewrite how different forms of experience are constructed, indexed, retrieved and routed.
Why this paper
From Mila – Quebec AI Institute and Université de Montréal
In one line
MemCodex improves task success by 10.1% with 3.4x fewer tokens and 2.1x faster inference.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ·No stated limitations found
- ✓Reports numbers on named benchmarks
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§