We present ACCORD, a framework and benchmark suite for disentangling the commonsense grounding and reasoning abilities of large language models (LLMs) through controlled, multi-hop counterfactuals. ACCORD introduces formal elements to commonsense reasoning to explicitly control and quantify reasoning complexity beyond the typical 1 or 2 hops. Uniquely, ACCORD can automatically generate benchmarks of arbitrary reasoning complexity, and so it scales with future LLM improvements. Benchmarking state-of-the-art LLMs -- including GPT-4o (2024-05-13), Llama-3-70B-Instruct, and Mixtral-8x22B-Instruct-v0.1 -- shows performance degrading to random chance with only moderate scaling, leaving substantial headroom for improvement. We release a leaderboard of the benchmark suite tested in this work, as well as code for automatically generating more complex benchmarks.
@article{arxiv.2406.02804,
title = {ACCORD: Closing the Commonsense Measurability Gap},
author = {François Roewer-Després and Jinyue Feng and Zining Zhu and Frank Rudzicz},
journal= {arXiv preprint arXiv:2406.02804},
year = {2025}
}
Comments
For leaderboard and dataset download, see https://www.codabench.org/competitions/3160/ For source code, see https://github.com/francois-rd/accord/