AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

Revela: Dense Retriever Learning via Language Modeling

Fengyu Cai, Tong Chen, Xinran Zhao, Sihao Chen, Hongming Zhang, Tongshuang Wu, Iryna Gurevych, Heinz Koeppl

ICLR 2026 · Large language models › Language, safety and interpretability · approach: LLM pruning and scaling data · 2 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
co-train retriever with LM via in-batch attention instead of REPLUG frozen-LM distillation (ablation)Revela 3B · CoIRnDCG@10 (%): 53.9 -> 60.1Improved
co-train retriever with LM instead of REPLUG (ablation)Revela 3B · BRIGHTnDCG@10 (%): 12.2 -> 20.1Improved
co-train with LM instead of REPLUG (ablation)Revela 1B · BEIRnDCG@10 (%): 35.7 -> 42.7Improved
use Revela instead of Contriever objective (wiki-trained, same LLaMA-3.2-1B) (ablation)Revela 1B · BEIRnDCG@10 (%): 42.4 -> 42.7No clear effect
use Revela instead of Contriever objective (wiki-trained) (ablation)Revela 1B · CoIRnDCG@10 (%): 50.3 -> 53.2Improved
use Revela instead of Contriever objective (code-trained, out-of-domain BEIR) (ablation)Revela 1B · BEIRnDCG@10 (%): 32.3 -> 39.6Improved
use Revela instead of Contriever objective (code-trained) (ablation)Revela 1B · CoIRnDCG@10 (%): 52.1 -> 58.6Improved
increase batch size from 4 to 16 (ablation)Revela 1B · CoIRnDCG@10 (%): 55.9 -> 58.6Improved
increase batch size from 4 to 16 (ablation)Revela 0.1B · CoIRnDCG@10 (%): 43.2 -> 48.4Improved
increase batch size from 4 to 16 (ablation)Revela 0.5B · BEIRnDCG@10 (%): 36.8 -> 38.8Improved
scale retriever backbone from 0.1B to 3B (ablation)Revela · CoIRnDCG@10 (%): 48.4 -> 60.1Improved
use a larger LM (1B instead of 0.1B) with 0.5B retriever (ablation)Revela 0.5B · BEIRnDCG@10 (%): 41.6 -> 38.8Hurt

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