AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

ExpWeaver: LLM Agents Learn from Experience via Latent RAG

Tao Feng, Tianyang Luo, Jingjun Xu, Zhigang Hua, Yan Xie, Shuang Yang, Ge Liu, Jiaxuan You

ICML 2026 · Large language models › Retrieval and knowledge · approach: Retrieval-augmented reasoning · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use latent retrieval and cross-attention instead of text-based experience retrieval (Search-R1)ExpWeaver (Qwen2.5-3B) · ExpBench-Genericaverage score: 73.27 -> 78.25Improved
use latent retrieval instead of Search-R1 text retrievalExpWeaver · Chem-TDCexact match: 63.78 -> 69.58Improved
use latent retrieval instead of text retrievalExpWeaver · ExpBench-Rec averageNDCG@10: 28.53 -> 44.40 (vs Search-R1)Improved
latent experience instead of concatenated textExpWeaver · token cost per querytokensImproved
train model on Chem-TDC with transferred experience bank vs zero-shot transferExpWeaver · Chem-TDC transferaccuracy: outperforms Search-R1 by over 8% in ZeroShot and about 8% in FewShotImproved
increase retrieval number K from 1 to 3 (ablation)ExpWeaver · five task categoriesperformanceImproved
increase retrieval number K from 3 to 8 (ablation)ExpWeaver · five task categoriesperformanceHurt
replace cross-attention with mean pooling (ablation)ExpWeaver · five task categoriesperformanceHurt
replace cross-attention with similarity-weighted mean (ablation)ExpWeaver · five task categoriesperformanceHurt
replace single-query cross-attention with Qformer (ablation)ExpWeaver · five task categoriesperformanceNo clear effect
set initial mixing coefficient lower bound a_min to 0.98 rather than 0.95 or 0.99 (ablation)ExpWeaver · five task categoriesperformanceImproved

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