YESciEval: Robust LLM-as-a-Judge for Scientific Question Answering

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: D'Souza, Jennifer, Giglou, Hamed Babaei, Münch, Quentin
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913865770467328
author D'Souza, Jennifer
Giglou, Hamed Babaei
Münch, Quentin
author_facet D'Souza, Jennifer
Giglou, Hamed Babaei
Münch, Quentin
contents Large Language Models (LLMs) drive scientific question-answering on modern search engines, yet their evaluation robustness remains underexplored. We introduce YESciEval, an open-source framework that combines fine-grained rubric-based assessment with reinforcement learning to mitigate optimism bias in LLM evaluators. We release multidisciplinary scienceQ&A datasets, including adversarial variants, with evaluation scores from multiple LLMs. Independent of proprietary models and human feedback, our approach enables scalable, cost-free evaluation. By advancing reliable LLM-as-a-judge models, this work supports AI alignment and fosters robust, transparent evaluation essential for scientific inquiry.
format Preprint
id arxiv_https___arxiv_org_abs_2505_14279
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle YESciEval: Robust LLM-as-a-Judge for Scientific Question Answering
D'Souza, Jennifer
Giglou, Hamed Babaei
Münch, Quentin
Computation and Language
Artificial Intelligence
Large Language Models (LLMs) drive scientific question-answering on modern search engines, yet their evaluation robustness remains underexplored. We introduce YESciEval, an open-source framework that combines fine-grained rubric-based assessment with reinforcement learning to mitigate optimism bias in LLM evaluators. We release multidisciplinary scienceQ&A datasets, including adversarial variants, with evaluation scores from multiple LLMs. Independent of proprietary models and human feedback, our approach enables scalable, cost-free evaluation. By advancing reliable LLM-as-a-judge models, this work supports AI alignment and fosters robust, transparent evaluation essential for scientific inquiry.
title YESciEval: Robust LLM-as-a-Judge for Scientific Question Answering
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2505.14279