Meta FAIR Self-Improving Memory Hits 56.0 on Continual Learning Bench

Meta FAIR researchers have released a method called Self-Improving Memory. The technique allows AI agents to write their own instructions for managing…

By Vane November 19, 2026 3 min read

Meta FAIR researchers have released a method called Self-Improving Memory. The technique allows AI agents to write their own instructions for managing file-based memory. A pilot study found that the AI-generated prompts achieved an average score of 56.0 on the Continual Learning Bench, beating the best human-written strategies which scored 44.6.

The core setup

The project uses the base model Kimi-K2.6. This is a 1T-parameter MoE with 32B active parameters and a 256K context window. The team added a LoRA adapter with a rank of 64 to the attention projection layers. Vast.ai suggests an 8x H200 node for serving Kimi-K2.6, though native INT4 weights are available.

The method relies on two prompts: LOAD and OFFLOAD. LOAD retrieves notes before a subtask begins. OFFLOAD updates the memory directory after the task. The system starts with four human-written seeds: Summary, Retrieval, Skill, and Hypothesis. The meta agent rewrites these prompts by running the full task stream five times for each candidate. The top four results by mean score become parents for the next revision cycle.

How distillation works

Stage two follows a Discover-Demonstrate-Distill pipeline. The task agent runs subtasks using the optimized prompts. Every model call is logged. The research team then swaps the optimized prompts back to generic ones in that history. The model is fine-tuned with LoRA rank 64 on attention projection layers. The loss covers only assistant tokens, including reasoning and tool calls.

The meta agent and task agent are the same Kimi-K2.6 model. The model is trained on memory skills it discovered itself. This works because the behavior stays valid under generic instructions. Examples include keeping QUICKSTART.md, REGISTRY.md and SCHEMA.md files, and verifying join keys before aggregation.

Performance results

The tests use three tasks from the Continual Learning Bench: Database (SQL), Codebase (repo fixes) and Spectrum (radio scans). Test subtasks follow a designed phase shift. This measures generalization to new instances of the same tasks.

Score breakdown

The table below shows the test scores. Kimi-K2.6 is used for all rows.

Test scoreDatabaseCodebaseSpectrumAverage
No memory7.531.821.920.4
Compaction39.836.934.637.1
Generic harness32.333.542.536.1
Best human-written47.943.742.144.6
Optimized harness53.864.349.956.0
SFT on generic trajectories31.637.543.637.6
SFT on optimized trajectories (generic prompts)37.745.146.543.1

Source: Figures 4 and 5, Self-Improving Memory blog.

Optimized harnesses beat the best hand-written seed on all three tasks. Codebase gained the most, at 20.6 points. Search was noisy: 37 of 44 generated revisions failed to beat the previous best. Database and Spectrum peaked in the first generation. Codebase kept improving through generation 4.

After distillation, every model runs the generic harness. The distilled model averages 43.1 across 10 matched seeds. SFT on generic trajectories reaches 37.6. That is a 5.5-point gain, or 14.6%.

Comparison with other methods

Several recent methods search over agent memory or harnesses. Benchmarks differ, so headline numbers are not directly comparable.

FeatureSelf-Improving MemoryALMAMemEvolveHarness-Zero
OrgFAIR at Meta, UW, JHUXiong, Hu, CluneOPPO AI Agent Team, LV-NUSPeking University, Google, HKUST
What is searchedLOAD/OFFLOAD memory promptsMemory designs as Python codeMemory architecture and contentsFull agent harness
Trains weightsYes, LoRA rank 64NoNoYes, LoRA SFT
ModelsKimi-K2.6 (agent and meta agent)GPT-5 meta agent, GPT-5-nano/mini task agentGPT-5-mini (plus DeepSeek V3.2, Kimi K2)Qwen3.5-9B student
BenchmarksContinual Learning Bench (3 tasks)ALFWorld, TextWorld, Baba Is AI, MiniHackGAIA, WebWalkerQA, xBench-DeepSearch, TaskCraftSpreadsheetBench Verified, AppWorld, USPTO Retrosynthesis
Headline result56.0 vs 44.6 best human harness; 43.1 distilled53.9% vs 41.1% no memory (GPT-5-mini)Up to 17.06% gains23.3% to 44.3% with harness removed
CodeNot releasedGitHubGitHubGitHub

Harness-Zero is the closest match, but it uses a separate agent to correct the student. Self-Improving Memory skips that step.

What it means

The research shows that an AI can improve its own workflow by rewriting its memory instructions. This is an internal research development. It does not imply changes to customer products, accounts, privacy or API behaviour. The team has not released the code.

Scroll to Top