paper-with-me

홈 › Papers

Do These LLM Benchmarks Agree? Fixing Benchmark Evaluation with BenchBench

2024-07-18 · Yotam Perlitz, Ariel Gera, Ofir Arviv, Asaf Yehudai, Elron Bandel, Eyal Shnarch, Michal Shmueli-Scheuer, Leshem Choshen

Recent advancements in Language Models (LMs) have catalyzed the creation of multiple benchmarks, designed to assess these models' general capabilities. A crucial task, however, is assessing the validity of the benchmarks themselves. This is most commonly done via Benchmark Agreement Testing (BAT), where new benchmarks are validated against established ones using some agreement metric (e.g., rank correlation). Despite the crucial role of BAT for benchmark builders and consumers, there are no standardized procedures for such agreement testing. This deficiency can lead to invalid conclusions, fostering mistrust in benchmarks and upending the ability to properly choose the appropriate benchmark to use. By analyzing over 40 prominent benchmarks, we demonstrate how some overlooked methodological choices can significantly influence BAT results, potentially undermining the validity of conclusions. To address these inconsistencies, we propose a set of best practices for BAT and demonstrate how utilizing these methodologies greatly improves BAT robustness and validity. To foster adoption and facilitate future research,, we introduce BenchBench, a python package for BAT, and release the BenchBench-leaderboard, a meta-benchmark designed to evaluate benchmarks using their peers. Our findings underscore the necessity for standardized BAT, ensuring the robustness and validity of benchmark evaluations in the evolving landscape of language model research. BenchBench Package: github.com/IBM/BenchBench Leaderboard: hf.co/spaces/IBM/BenchBench

📄 PDF Abstract BibTeX arXiv:2407.13696

Code (1)

ibm/benchbench 공식 구현

Tasks

Language Modelling

Methods 이 논문이 사용한 방법론

SET Dynamic Sparse Training method where weight mask is updated randomly periodically

Similar Papers 제목 키워드 기반

Let's Agree to Agree: Neural Networks Share Classification Order on Real Datasets

2019-05-26 · ICML 2020 1 · Guy Hacohen, Leshem Choshen, Daphna Weinshall

We report a series of robust empirical observations, demonstrating that deep Neural Networks learn the examples in both the training and test sets in a similar order. This phenomenon is observed in all the commonly used …

General Classificationimage-classificationImage Classificationtext-classification+1

The Fault in our Stars: Quality Assessment of Code Generation Benchmarks

2024-04-15 · Mohammed Latif Siddiq, Simantika Dristi, Joy Saha, Joanna C. S. Santos

Large Language Models (LLMs) are gaining popularity among software engineers. A crucial aspect of developing effective code generation LLMs is to evaluate these models using a robust benchmark. Evaluation benchmarks with…

Code GenerationMemorization

Diff-Based Code Corruption using LLMs for Large-Scale Bugfix Benchmarking

2026-06-27 · Balázs Szalontai, Ábel Szauter, Balázs Márton, Péter Verebics 외 arxiv

There are various benchmarks to evaluate bugfixing capabilities of Large Language Models. However, most widespread benchmarks do not fully reflect real-world bugfixing practices. They are small, weakening statistical rel…

PerfBench: Can Agents Resolve Real-World Performance Bugs?

2025-09-28 · Spandan Garg, Roshanak Zilouchian Moghaddam, Neel Sundaresan arxiv

Performance bugs are inefficiencies in software that waste computational resources without causing functional failures, making them particularly challenging to detect and fix. While recent advances in Software Engineerin…

ProjectTest: A Project-level LLM Unit Test Generation Benchmark and Impact of Error Fixing Mechanisms

2025-02-10 · Yibo Wang, Congying Xia, Wenting Zhao, Jiangshu Du 외

Unit test generation has become a promising and important use case of LLMs. However, existing evaluation benchmarks for assessing LLM unit test generation capabilities focus on function- or class-level code rather than m…