ARB: Advanced Reasoning Benchmark for Large Language Models
This addresses the need for more rigorous evaluation of LLMs in advanced reasoning for researchers and developers, though it is incremental as it builds on prior benchmarking efforts.
The authors tackled the problem of existing benchmarks becoming less useful as LLMs achieve high scores, by introducing ARB, a more challenging benchmark for advanced reasoning across multiple fields, and found that current models like GPT-4 score below 50% on demanding tasks.
Large Language Models (LLMs) have demonstrated remarkable performance on various quantitative reasoning and knowledge benchmarks. However, many of these benchmarks are losing utility as LLMs get increasingly high scores, despite not yet reaching expert performance in these domains. We introduce ARB, a novel benchmark composed of advanced reasoning problems in multiple fields. ARB presents a more challenging test than prior benchmarks, featuring problems in mathematics, physics, biology, chemistry, and law. As a subset of ARB, we introduce a challenging set of math and physics problems which require advanced symbolic reasoning and domain knowledge. We evaluate recent models such as GPT-4 and Claude on ARB and demonstrate that current models score well below 50% on more demanding tasks. In order to improve both automatic and assisted evaluation capabilities, we introduce a rubric-based evaluation approach, allowing GPT-4 to score its own intermediate reasoning steps. Further, we conduct a human evaluation of the symbolic subset of ARB, finding promising agreement between annotators and GPT-4 rubric evaluation scores.