In this work, we introduce a novel evaluation paradigm for Large Language Models (LLMs) that compels them to transition from a traditional question-answering role, akin to a student, to a solution-scoring role, akin to a teacher. This paradigm, focusing on "reasoning about reasoning," hence termed meta-reasoning, shifts the emphasis from result-oriented assessments, which often neglect the reasoning process, to a more comprehensive evaluation that effectively distinguishes between the cognitive capabilities of different models. By applying this paradigm in the GSM8K dataset, we have developed the MR-GSM8K benchmark. Our extensive analysis includes several state-of-the-art models from both open-source and commercial domains, uncovering fundamental deficiencies in their training and evaluation methodologies. Notably, while models like Deepseek-v2 and Claude3-Sonnet closely competed with GPT-4 in GSM8K, their performance disparities expanded dramatically in MR-GSM8K, with differences widening to over 20 absolute points, underscoring the significant challenge posed by our meta-reasoning approach.
MR-GSM8K: A Meta-Reasoning Benchmark for Large Language Model Evaluation
A novel evaluation paradigm for Large Language Models focuses on meta-reasoning to uncover cognitive deficiencies, offering a more accurate assessment than traditional benchmarks.
- Year
- 2023
- Venue
- arXiv 2023
- Authors
- 5
- Hosting
- Abstract onlyARXIV-DEFAULT
Cite
Notes
Only stored in your browser.
Attribution
- Abstract & full text
- arxiv.org/abs/2312.17080v4ARXIV-DEFAULT
- TL;DR
- Semantic Scholar