AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

RAS: Retrieval-And-Structuring for Knowledge-Intensive LLM Generation

Pengcheng Jiang, Lang Cao, Ruike Zhu, Minhao Jiang, Yunyi Zhang, Jiaming Shen, Jimeng Sun, Jiawei Han

ICLR 2026 · Large language models › Retrieval and knowledge · approach: RAG robustness and evaluation · 19 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
add iterative planning and retrieval (ablation)RAS 7B · TriviaQAaccuracy: +8.8%Improved
add iterative planning and retrieval (ablation)RAS 7B · 2WikiMultihopQAF1: +9.0%Improved
add text-to-triple conversion (ablation)RAS 7B · 2WikiMultihopQAF1: -9.0% withoutImproved
add text-to-triple conversion (ablation)RAS 7B · ASQAMAUVE: -22.2% withoutImproved
add graph encoder (GraphToken) in training (ablation)RAS 7B · PubHealthaccuracy: -11.2% withoutImproved
add graph encoder (GraphToken) at inference (ablation)RAS 7B · PubHealthaccuracy: -10.0% withoutImproved
add LoRA to graph token tuning (ablation)RAS 7B · average across benchmarksaccuracy: -11.8% on average without LoRAImproved
add multi-task training of planning and answering (ablation)RAS 7B · PubHealthaccuracy: -12.3% withoutImproved
use higher-quality triple extractor (Claude-3.5-Sonnet vs LLaMA-3.2-3B) (ablation)RAS 7B · TriviaQAaccuracy: 72.7 -> 73.8Improved
use Flan-T5-Large instead of LLaMA-3.2-3B triple extractor (ablation)RAS 7B · TriviaQAaccuracy: 72.7 -> 70.3Hurt
increase fraction of graph triples retained (10% to 100%) (ablation)RAS 7B/8B · TriviaQAaccuracyImproved
increase training data volume (ablation)RAS 7B · TQA, 2WQA, ELI5accuracy/rougeMixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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