AI Paper Atlas

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

ElicitR: Unlocking Latent Reasoning in Dense Retrievers via Generative Regularization

Fengyu Cai, Iryna Gurevych, Heinz Koeppl

ICML 2026 · Large language models › Language, safety and interpretability · approach: Faithfulness of LLM explanations · 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
add generative regularization (ElicitR) to contrastive trainingRepLLaMA, 0.1B · BRIGHT avgnDCG@10 (%): 12.2 -> 15.7Improved
add generative regularization (ElicitR) to contrastive trainingRepLLaMA, 0.5B · BRIGHT avgnDCG@10 (%): 16.4 -> 19.0Improved
add generative regularization (ElicitR) to contrastive trainingRepLLaMA, 1B · BRIGHT avgnDCG@10 (%): 17.7 -> 21.4Improved
add generative regularization (ElicitR) to contrastive trainingRepLLaMA, 3B · BRIGHT avgnDCG@10 (%): 20.0 -> 23.1Improved
add generative regularization (ElicitR) to contrastive trainingRepLLaMA, 3B · BRIGHT Pony and AoPSnDCG@10 (%): 9.3 -> 3.0 (Pony); 14.8 -> 12.8 (AoPS)Hurt
add generative regularization (ElicitR) to contrastive trainingRepLLaMA, 3B · BEIR avgnDCG@10 (%): 55.9 -> 55.4No clear effect
add generative regularization (ElicitR) to contrastive trainingRepLLaMA, 1B · BEIR avgnDCG@10 (%): 53.6 -> 53.8No clear effect
add raw-text initialization of retriever and LM before MS MARCO co-training (ablation)ElicitR, 1B and 3B · BRIGHT avgnDCG@10 (%): 18.6 -> 21.4 (1B); 20.3 -> 23.1 (3B)Improved
initialize the retriever as well as the LM (instead of LM only) (ablation)ElicitR, 1B and 3B · BRIGHT avgnDCG@10 (%): 19.3 -> 21.4 (1B); 20.9 -> 23.1 (3B)Improved
keep the reference LM trainable instead of frozen (ablation)ElicitR, 1B and 3B · BRIGHT avgnDCG@10 (%): 21.3 -> 21.4 (1B); 22.2 -> 23.1 (3B)Improved
add MS MARCO contrastive fine-tuning on top of raw-text-initialized ElicitR (ablation)ElicitR-init · BRIGHT avgnDCG@10 (%): 13.8 -> 21.4 (1B); 15.8 -> 23.1 (3B)Improved
use Wikipedia-only instead of mixed Wikipedia and code initialization (ablation)ElicitR, 1B and 3B · BRIGHT avgnDCG@10 (%): 21.4 -> 21.9 (1B); 23.1 -> 23.0 (3B)No clear effect

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