AI Paper Atlas

Research map / Large language models / Reasoning and chain of thought

QuRL: Rubrics As Judge For Open-Ended Question Answering

Xiyu Wei, Qingwei Zong, Xiaoguang Li, Eugene J. Yu, Sujian Li

ICLR 2026 · Large language models › Reasoning and chain of thought · approach: RL with verifiable rewards for reasoning

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes QuRL, an RL approach for open-ended question answering that uses rubrics as a judge to supply reward signals in place of verifiable ground truth.

Problem

RLVR works well for tasks with gold answers (code, math) but open-ended QA lacks reliable evaluation and verifiable reward signals.

Approach

Uses rubrics as the judge for open-ended QA so that RL can be applied where no single verifiable answer exists; the abstract only indicates rubric-based rewards replace gold-truth verification.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
GRPO with case-wise rubric reward instead of SFT on R1 responsesQwen2.5-7B · Average of 3 benchmarksnormalized score: 42.3 -> 59.3Improved
use case-wise rubric reward instead of RLHF scalar reward model (ablation)QuRL · Average of 3 benchmarksnormalized score: 47.7 -> 59.3Improved
add case-wise rubrics (vs five-dimension prompt reward) (ablation)QuRL · Average of 3 benchmarksnormalized score: 44.0 -> 59.3Improved
add rubric filter (ablation)QuRL · Average of 3 benchmarksnormalized score: 52.2 -> 59.3Improved
add web-derived meta-descriptions to rubric generation (ablation)QuRL · Average of 3 benchmarksnormalized score: 48.9 -> 59.3Improved
use fixed-maximum rubric scoring instead of free-form (ablation)QuRL · Average of 3 benchmarksnormalized score: 50.4 -> 59.3Improved
use fixed-maximum scoring instead of judge-based binary scoring (ablation)QuRL · Average of 3 benchmarksnormalized score: 52.4 -> 59.3Improved
score with QuRL rubrics instead of plain LLM-as-a-JudgeGPT-4o judge · Human consistencySpearman rho: 0.08 -> 0.31Improved
give LLM judges web articles as referencesGPT-4o and DeepSeek-R1 judges · 50 sampled questionsSpearman rho vs human: 0.139 -> 0.209 (GPT-4o w/ web articles); 0.081 -> 0.170 (DeepSeek-R1)Improved
add QuRL rubrics for GPT-4o judgeGPT-4o judge · 50 sampled questionsSpearman rho vs human: 0.139 -> 0.260Improved

Experimental setup

Builds on: RLVR

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