Testing Uncertainty of Large Language Models for Physics Knowledge and Reasoning
원문 초록
Large Language Models (LLMs) have gained significant popularity in recent years for their ability to answer questions in various fields. However, these models have a tendency to "hallucinate" their responses, making it challenging to evaluate their performance. A major challenge is determining how to assess the certainty of a model's predictions and how it correlates with accuracy. In this work, we introduce an analysis for evaluating the performance of popular open-source LLMs, as well as gpt-3.5 Turbo, on multiple choice physics questionnaires. We focus on the relationship between answer accuracy and variability in topics related to physics. Our findings suggest that most models provide accurate replies in cases where they are certain, but this is by far not a general behavior. The relationship between accuracy and uncertainty exposes a broad horizontal bell-shaped distribution. We report how the asymmetry between accuracy and uncertainty intensifies as the questions demand more logical reasoning of the LLM agent, while the same relationship remains sharp for knowledge retrieval tasks.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
LogicOCR: Do Your Large Multimodal Models Excel at Logical Reasoning on Text-Rich Images?
Exploring the Effect of Reinforcement Learning on Video Understanding: Insights from SEED-Bench-R1
SCoRE: Benchmarking Long-Chain Reasoning in Commonsense Scenarios
AutoLogi: Automated Generation of Logic Puzzles for Evaluating Reasoning Abilities of Large Language Models
CP-Router: An Uncertainty-Aware Router Between LLM and LRM
Reversal Blessing: Thinking Backward May Outpace Thinking Forward in Multi-choice Questions
토론
이 논문과 연결된 포럼 글이 아직 없습니다.