AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

TG-RAG: A Retrieval-Augmented Framework for Reasoning Guidance in Specialized Domains

Liang Su, Mingyang Zhang, Yun Xiong, Tengfei LIU, Siwei Zhang, Xi Chen, Li Sun

ICML 2026 · Large language models › Retrieval and knowledge · approach: Retrieval-augmented reasoning

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 Chain-of-Trees EPG instead of fixed-path EPG (ablation)TG-RAG (R1-Distill-32B) · Financeaccuracy: 0.76 -> 0.81Improved
use Chain-of-Trees EPG instead of fixed-path EPG (ablation)TG-RAG (R1-Distill-32B) · Medicalaccuracy: 0.86 -> 0.92Improved
use Chain-of-Trees EPG instead of fixed-path EPG (ablation)TG-RAG (R1-Distill-32B) · Astronomyaccuracy: 0.80 -> 0.90Improved
inject guidance into reasoning process instead of in-prompt (ablation)TG-RAG (R1-Distill-32B) · Financeaccuracy: 0.72 -> 0.81Improved
inject guidance into reasoning process instead of in-prompt (ablation)TG-RAG (R1-Distill-32B) · Medicalaccuracy: 0.90 -> 0.92Improved
inject guidance into reasoning process instead of in-prompt (ablation)TG-RAG (R1-Distill-32B) · Astronomyaccuracy: 0.78 -> 0.90Improved
use dynamic IRG injection instead of static template prompting (ablation)TG-RAG (R1-Distill-32B) · Financeaccuracy: 0.67 -> 0.81Improved
use dynamic IRG injection instead of static template prompting (ablation)TG-RAG (R1-Distill-32B) · Medicalaccuracy: 0.86 -> 0.92Improved
use dynamic IRG injection instead of static template prompting (ablation)TG-RAG (R1-Distill-32B) · Astronomyaccuracy: 0.80 -> 0.90Improved
add both knowledge and guidance pathways vs Only Knowledge/Only Guidance (ablation)TG-RAG (Qwen3-32B) · Financeaccuracy: 51.0 -> 76.0Improved
add both knowledge and guidance pathways vs Only Knowledge/Only Guidance (ablation)TG-RAG (Qwen3-32B) · Astronomyaccuracy: 90.0 -> 92.0No clear effect
use TG-RAG instead of Few-CoTR1-Distill-32B · GSM8Kaccuracy: 97.2 -> 96.1Hurt

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