AbGen: Evaluating Large Language Models in Ablation Study Design and Evaluation for Scientific Research

• URL: http://arxiv.org/abs/2507.13300v1
• Date: Thu, 17 Jul 2025 17:09:22 GMT
• Title: AbGen: Evaluating Large Language Models in Ablation Study Design and Evaluation for Scientific Research
• Authors: Yilun Zhao, Weiyuan Chen, Zhijian Xu, Manasi Patwardhan, Yixin Liu, Chengye Wang, Lovekesh Vig, Arman Cohan,
• Abstract summary: AbGen is the first benchmark designed to evaluate the capabilities of LLMs in designing ablation studies for scientific research.<n>To better investigate this, we develop AbGen-Eval, a meta-evaluation benchmark designed to assess the reliability of commonly used automated evaluation systems.
• Score: 33.79419161415481
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We introduce AbGen, the first benchmark designed to evaluate the capabilities of LLMs in designing ablation studies for scientific research. AbGen consists of 1,500 expert-annotated examples derived from 807 NLP papers. In this benchmark, LLMs are tasked with generating detailed ablation study designs for a specified module or process based on the given research context. Our evaluation of leading LLMs, such as DeepSeek-R1-0528 and o4-mini, highlights a significant performance gap between these models and human experts in terms of the importance, faithfulness, and soundness of the ablation study designs. Moreover, we demonstrate that current automated evaluation methods are not reliable for our task, as they show a significant discrepancy when compared to human assessment. To better investigate this, we develop AbGen-Eval, a meta-evaluation benchmark designed to assess the reliability of commonly used automated evaluation systems in measuring LLM performance on our task. We investigate various LLM-as-Judge systems on AbGen-Eval, providing insights for future research on developing more effective and reliable LLM-based evaluation systems for complex scientific tasks.

Related papers

• ResearchBench: Benchmarking LLMs in Scientific Discovery via Inspiration-Based Task Decomposition [67.26124739345332]
  Large language models (LLMs) have demonstrated potential in assisting scientific research, yet their ability to discover high-quality research hypotheses remains unexamined.<n>We introduce the first large-scale benchmark for evaluating LLMs with a near-sufficient set of sub-tasks of scientific discovery.<n>We develop an automated framework that extracts critical components - research questions, background surveys, inspirations, and hypotheses - from scientific papers.
  arXiv  Detail & Related papers  (2025-03-27T08:09:15Z)
• FutureGen: LLM-RAG Approach to Generate the Future Work of Scientific Article [6.682911432177815]
  This study generates future work suggestions from key sections of a scientific article alongside related papers.<n>We experimented with various Large Language Models (LLMs) and integrated Retrieval-Augmented Generation (RAG) to enhance the generation process.
  arXiv  Detail & Related papers  (2025-03-20T06:14:02Z)
• MME-Survey: A Comprehensive Survey on Evaluation of Multimodal LLMs [97.94579295913606]
  Multimodal Large Language Models (MLLMs) have garnered increased attention from both industry and academia.<n>In the development process, evaluation is critical since it provides intuitive feedback and guidance on improving models.<n>This work aims to offer researchers an easy grasp of how to effectively evaluate MLLMs according to different needs and to inspire better evaluation methods.
  arXiv  Detail & Related papers  (2024-11-22T18:59:54Z)
• IdeaBench: Benchmarking Large Language Models for Research Idea Generation [19.66218274796796]
  Large Language Models (LLMs) have transformed how people interact with artificial intelligence (AI) systems. We propose IdeaBench, a benchmark system that includes a comprehensive dataset and an evaluation framework. Our dataset comprises titles and abstracts from a diverse range of influential papers, along with their referenced works. Our evaluation framework is a two-stage process: first, using GPT-4o to rank ideas based on user-specified quality indicators such as novelty and feasibility, enabling scalable personalization.
  arXiv  Detail & Related papers  (2024-10-31T17:04:59Z)
• Unveiling and Consulting Core Experts in Retrieval-Augmented MoE-based LLMs [64.9693406713216]
  Internal mechanisms that contribute to the effectiveness of RAG systems remain underexplored. Our experiments reveal that several core groups of experts are primarily responsible for RAG-related behaviors. We propose several strategies to enhance RAG's efficiency and effectiveness through expert activation.
  arXiv  Detail & Related papers  (2024-10-20T16:08:54Z)
• LLaVA-Critic: Learning to Evaluate Multimodal Models [110.06665155812162]
  We introduce LLaVA-Critic, the first open-source large multimodal model (LMM) designed as a generalist evaluator.<n>LLaVA-Critic is trained using a high-quality critic instruction-following dataset that incorporates diverse evaluation criteria and scenarios.
  arXiv  Detail & Related papers  (2024-10-03T17:36:33Z)
• From Text to Insight: Leveraging Large Language Models for Performance Evaluation in Management [6.70908766695241]
  This study explores the potential of Large Language Models (LLMs), specifically GPT-4, to enhance objectivity in organizational task performance evaluations. Our results suggest that GPT ratings are comparable to human ratings but exhibit higher consistency and reliability. Our research suggests that while LLMs are capable of extracting meaningful constructs from text-based data, their scope is currently limited to specific forms of performance evaluation.
  arXiv  Detail & Related papers  (2024-08-09T20:35:10Z)
• HumanEvo: An Evolution-aware Benchmark for More Realistic Evaluation of Repository-level Code Generation [36.1669124651617]
  We conduct an empirical study to understand Large Language Models' code generation performance within settings that reflect the evolution nature of software development.<n>We use an evolution-aware repository-level code generation dataset, namely HumanEvo, equipped with an automated execution-based evaluation tool.<n>We find that previous evolution-ignored evaluation methods result in inflated performance of LLMs, with performance overestimations ranging from 10.0% to 61.1%.
  arXiv  Detail & Related papers  (2024-06-11T03:19:18Z)
• Comprehensive Reassessment of Large-Scale Evaluation Outcomes in LLMs: A Multifaceted Statistical Approach [64.42462708687921]
  Evaluations have revealed that factors such as scaling, training types, architectures and other factors profoundly impact the performance of LLMs. Our study embarks on a thorough re-examination of these LLMs, targeting the inadequacies in current evaluation methods. This includes the application of ANOVA, Tukey HSD tests, GAMM, and clustering technique.
  arXiv  Detail & Related papers  (2024-03-22T14:47:35Z)
• Evaluating the Performance of Large Language Models on GAOKAO Benchmark [53.663757126289795]
  This paper introduces GAOKAO-Bench, an intuitive benchmark that employs questions from the Chinese GAOKAO examination as test samples. With human evaluation, we obtain the converted total score of LLMs, including GPT-4, ChatGPT and ERNIE-Bot. We also use LLMs to grade the subjective questions, and find that model scores achieve a moderate level of consistency with human scores.
  arXiv  Detail & Related papers  (2023-05-21T14:39:28Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.