CLAIJun 20, 2024

MR-Ben: A Meta-Reasoning Benchmark for Evaluating System-2 Thinking in LLMs

arXiv:2406.13975v345 citationsHas Code
Originality Incremental advance
AI Analysis

This addresses the problem of saturated outcome-based benchmarks for AI researchers by providing a process-based evaluation tool, though it is incremental as it builds on existing reasoning assessment methods.

The authors tackled the challenge of evaluating reasoning abilities in large language models by introducing MR-Ben, a meta-reasoning benchmark with 5,975 questions across various subjects, which revealed significant performance gaps, such as many state-of-the-art models falling behind while models like OpenAI's o1 series showed strong results.

Large language models (LLMs) have shown increasing capability in problem-solving and decision-making, largely based on the step-by-step chain-of-thought reasoning processes. However, evaluating these reasoning abilities has become increasingly challenging. Existing outcome-based benchmarks are beginning to saturate, becoming less effective in tracking meaningful progress. To address this, we present a process-based benchmark MR-Ben that demands a meta-reasoning skill, where LMs are asked to locate and analyse potential errors in automatically generated reasoning steps. Our meta-reasoning paradigm is especially suited for system-2 slow thinking, mirroring the human cognitive process of carefully examining assumptions, conditions, calculations, and logic to identify mistakes.MR-Ben comprises 5,975 questions curated by human experts across a wide range of subjects, including physics, chemistry, logic, coding, and more. Through our designed metrics for assessing meta-reasoning on this benchmark, we identify interesting limitations and weaknesses of current LLMs (open-source and closed-source models). For example, with models like the o1 series from OpenAI demonstrating strong performance by effectively scrutinizing the solution space, many other state-of-the-art models fall significantly behind on MR-Ben, exposing potential shortcomings in their training strategies and inference methodologies.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes