AI Paper Atlas

Research map / Large language models / Instruction and fine-tuning

ERank: Fusing Supervised Fine-Tuning and Reinforcement Learning for Effective and Efficient Text Reranking

Yuzheng Cai, Yanzhao Zhang, Dingkun Long, Mingxin Li, Pengjun Xie, Weiguo Zheng

AAAI 2026 · Large language models › Instruction and fine-tuning · approach: Fine-tuning for reasoning data · 5 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
replace yes/no scoring with 0-10 integer scores (ablation)QwQ-32B reranker · BRIGHT, TREC DL, BEIR-5nDCG@10: BRIGHT 20.8 -> 23.2; TREC DL 60.9 -> 66.1; BEIR-5 31.1 -> 37.1Improved
use 0-10 scores instead of 0-3 scores (ablation)QwQ-32B reranker · BRIGHT, TREC DL, BEIR-5nDCG@10: 22.7 -> 23.2; 64.1 -> 66.1; 36.5 -> 37.1Improved
add RL stage after SFT (ablation)ERank-4B · BRIGHT, BEIR-5, TREC DLnDCG@10: BRIGHT 22.0 -> 22.7; BEIR-5 30.0 -> 32.4; TREC DL 68.1 -> 68.9; FollowIR 11.2 -> 11.0Mixed
add SFT before RL (ablation)ERank-4B · BRIGHT, TREC DLnDCG@10: RL only 20.1 -> 22.7 BRIGHT; 66.5 -> 68.9 TREC DLImproved
use listwise-derived reward r_RR instead of pointwise squared-error reward r_SE (ablation)ERank-4B RL · TREC DLnDCG@10: 62.6 -> 68.9Improved
use r_RR instead of r_nDCG (ablation)ERank-4B RL · BRIGHT, FollowIRnDCG@10 / p-MRR: BRIGHT 21.5 -> 22.7; FollowIR 13.2 -> 11.0Mixed
add BM25 hybrid score aggregationERank-4B · BRIGHT (ReasonIR-8B retrieval)nDCG@10: 30.5 -> 38.7Improved
scale ERank from 4B to 32BERank · BRIGHTnDCG@10: 22.7 -> 24.4 (BM25 original queries)Improved
use 3-sample consensus label selection vs single generation in data synthesis (ablation)teacher labels (QwQ-32B) · 512-query samplenDCG@10: 63.5 -> 65.6; 10 samples 67.4Improved

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