paper-with-me

Papers

MCQA-Eval: Efficient Confidence Evaluation in NLG with Gold-Standard Correctness Labels

2025-02-20 · Xiaoou Liu, Zhen Lin, Longchao Da, Chacha Chen, Shubhendu Trivedi, Hua Wei

Large Language Models (LLMs) require robust confidence estimation, particularly in critical domains like healthcare and law where unreliable outputs can lead to significant consequences. Despite much recent work in confidence estimation, current evaluation frameworks rely on correctness functions -- various heuristics that are often noisy, expensive, and possibly introduce systematic biases. These methodological weaknesses tend to distort evaluation metrics and thus the comparative ranking of confidence measures. We introduce MCQA-Eval, an evaluation framework for assessing confidence measures in Natural Language Generation (NLG) that eliminates dependence on an explicit correctness function by leveraging gold-standard correctness labels from multiple-choice datasets. MCQA-Eval enables systematic comparison of both internal state-based white-box (e.g. logit-based) and consistency-based black-box confidence measures, providing a unified evaluation methodology across different approaches. Through extensive experiments on multiple LLMs and widely used QA datasets, we report that MCQA-Eval provides efficient and more reliable assessments of confidence estimation methods than existing approaches.

📄 PDF Abstract BibTeX arXiv:2502.14268

Code (0)

등록된 구현이 없습니다.

Tasks

Multiple-choiceText Generation

Similar Papers 제목 키워드 기반

Do Bias Benchmarks Generalise? Evidence from Voice-based Evaluation of Gender Bias in SpeechLLMs

2025-09-24 · Shree Harsha Bokkahalli Satish, Gustav Eje Henter, Éva Székely arxiv

Recent work in benchmarking bias and fairness in speech large language models (SpeechLLMs) has relied heavily on multiple-choice question answering (MCQA) formats. The model is tasked to choose between stereotypical, ant…

Question Answering

Reassessing High-Performing LLMs on Polish Medical Exams: True Competence or Bias-Driven Performance?

2026-06-10 · Antoni Lasik, Jakub Pokrywka, Łukasz Grzybowski, Jeremi Ignacy Kaczmarek 외 arxiv

Large language models (LLMs) in medicine are mainly evaluated using multiple-choice question answering (MCQA), which can overestimate real clinical ability due to guessing strategies and answer biases. To address these l…

Question Answering

Are We Evaluating Knowledge or Phrasing? Mitigating MCQA Sensitivity with ParaEval

2026-06-09 · João Maria Janeiro, Mathurin Videau, Andrea Caciolai, Benjamin Piwowarski 외 arxiv

Multiple-choice (MCQA) benchmarks are the standard for evaluating pretrained large language models, but their reliance on log-likelihood scoring makes them unreliable. Specifically, standard scores are highly sensitive t…

Auditing MCQA Benchmarks through Probability Landscapes

2026-08-31 · Minsoo Song, Chanjun Park arxiv

As Large Language Models rapidly advance, performance on standard multiple-choice question answering (MCQA) benchmarks is reaching saturation. While the community has responded by developing increasingly difficult datase…

Question Answering

LLMs May Perform MCQA by Selecting the Least Incorrect Option

2024-02-02 · Haochun Wang, Sendong Zhao, Zewen Qiang, Nuwa Xi 외

In the field of NLP, Large Language Models (LLMs) have markedly enhanced performance across a variety of tasks. However, the comprehensive evaluation of LLMs remains an inevitable challenge for the community. Recently, t…

Multiple-choiceMultiple Choice Question Answering (MCQA)Question AnsweringText Generation