AI Paper Atlas

Research map / Multi-agent systems and games / LLM agents

From Interactions to Principles: Experience-Driven Self-Distillation for Evolving LLM Agents

Rong Wu, Xiaoman Wang, Jianbiao Mei, Pinlong Cai, Daocheng Fu, Cheng Yang, Licheng Wen, Xuemeng Yang, Yufan Shen, Yuxin Wang, Botian Shi

ICML 2026 · Multi-agent systems and games › LLM agents · approach: Self-evolving language agent learning

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes experience-driven self-distillation that lets an LLM agent turn its own interaction traces into reusable principles and distill them back into its policy.

Problem

LLM agents are largely stateless across episodes, discarding interaction traces; prior methods either store raw trajectories for case-based reuse or rely on external teacher models for reflections, which limits generalization or leaves the policy unchanged.

Approach

The agent distills reusable principles from its own experience and self-distills them into its policy, avoiding external teacher models and raw trajectory storage (based on the abstract only).

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add experience retrieval at inference (ablation)EvolveR (Qwen2.5-3B) · 7 QA benchmarksaverage exact match: 0.340 -> 0.382Improved
add experience retrieval at inference (ablation)EvolveR (Qwen2.5-0.5B) · 7 QA benchmarksaverage exact match: 0.078 -> 0.150Improved
add experience retrieval at inference (ablation)EvolveR (Qwen2.5-1.5B) · 7 QA benchmarksaverage exact match: 0.123 -> 0.270Improved
use own model for distillation instead of GPT-4o-mini (ablation)EvolveR (Qwen2.5-3B) · 7 QA benchmarksaverage exact match: 0.370 -> 0.382Improved
use own model for distillation instead of GPT-4o-mini (ablation)EvolveR (Qwen2.5-0.5B) · 7 QA benchmarksaverage exact match: 0.220 -> 0.150Hurt
use own model for distillation instead of GPT-4o-mini (ablation)EvolveR (Qwen2.5-1.5B) · 7 QA benchmarksaverage exact match: 0.290 -> 0.270Hurt
add RL training to retrieval with distilled principles (ablation)EvolveR (Qwen2.5-3B) · 7 QA benchmarksaverage exact match: 0.357 -> 0.382Improved
add RL training to a no-experience agent (ablation)Qwen2.5-3B · 7 QA benchmarksaverage exact match: 0.134 -> 0.325Improved
add principle distillation and retrieval to RL (ablation)Qwen2.5-3B · 7 QA benchmarksaverage exact match: 0.325 -> 0.382Improved
add principle retrieval to distilled principles with RL (ablation)Qwen2.5-3B · 7 QA benchmarksaverage exact match: 0.340 -> 0.382Improved
replace Search-R1-instruct with EvolveRQwen2.5-3B · Average of 7 QA benchmarksaverage exact match: 0.325 -> 0.382Improved
replace Search-R1-instruct with EvolveRQwen2.5-7B · Average of 7 QA benchmarksaverage exact match: 0.385 -> 0.417Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy