iCreatAI News
Skip to content
Trending storyUpdating

Meta FAIR proposes self-improving memory method

1 reports1 sources14 hr agoUpdated

Understand this story

AI overview

Meta FAIR, together with the University of Washington and Johns Hopkins University, proposed the Self-Improving Memory method, which lets AI agents search for LOAD/OFFLOAD memory prompts themselves and then distill them into Kimi-K2.6 weights using LoRA rank 64. The method reaches 56.0 on the Continual Learning Bench. According to Meta FAIR, this method targets AI agents that need to manage their own memory; the above benchmark results come from the team's published results.

AI overview based on source reports · Updated 13 hr ago

Coverage timeline

Follow the coverage to explore different perspectives.

Nov 20
  1. MarkTechPost
    Meta FAIR Self-Improving Memory Hits 56.0 on Continual Learning Bench

    Meta FAIR, with University of Washington and Johns Hopkins University researchers, introduced Self-Improving Memory, a two-stage method in which a Kimi-K2.6 meta agent rewrites LOAD and OFFLOAD memory prompts and the resulting skills are distilled into the model's weights via a rank-64 LoRA adapter on attention projections. Self-found memory prompts averaged 56.0 across three Continual Learning Bench tasks versus 44.6 for the best human-written prompts, with Codebase reaching 64.3 (+20.6), while the distilled model averaged 43.1 versus 37.6 for SFT on generic trajectories. The post notes that 37 of 44 generated prompt revisions failed to beat the previous best, and that code was not released.

Activity for this story

Current comparable activity
7
Peak comparable activity
10Nov 20 07:00
Change in the past 24 hours
–

Trends compare the same fully observed participants, which may cover fewer participants than current activity. Move the pointer or tap the chart to inspect hourly activity; use the left and right arrow keys with a keyboard.