{"paper":{"title":"CommonWhy: A Dataset for Evaluating Entity-Based Causal Commonsense Reasoning in Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"CommonWhy introduces 15,000 why questions that test whether LLMs can combine specific entity facts with causal commonsense inference","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Armin Toroghi, Faeze Moradi Kalarde, Scott Sanner","submitted_at":"2026-05-13T02:47:21Z","abstract_excerpt":"To effectively interact with the real world, Large Language Models (LLMs) require entity-based commonsense reasoning, a challenging task that necessitates integrating factual knowledge about specific entities with commonsense inference. Existing datasets for evaluating LLM entity-based commonsense reasoning have largely focused on True/False or multiple-choice questions, leaving the explicit assessment of the model's ability in abductive reasoning about causes and effects and generating explanations largely unexamined. In this work, we introduce CommonWhy, a dataset of 15,000 why questions des"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Experiments with state-of-the-art LLMs and LLM-based KGQA methods reveal their significant shortcomings, including frequent factual hallucinations and failures in causal reasoning.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The questions in CommonWhy require genuine integration of entity facts with causal commonsense reasoning rather than being solvable through superficial patterns learned during training.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"CommonWhy is a new dataset of 15,000 why-questions for evaluating LLMs on entity-based causal commonsense reasoning grounded in Wikidata.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"CommonWhy introduces 15,000 why questions that test whether LLMs can combine specific entity facts with causal commonsense inference","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"168942156c9252f35eef4703d0a82277946ad9fd4eb8f7b6c4486af0c4859e92"},"source":{"id":"2605.12918","kind":"arxiv","version":1},"verdict":{"id":"3898ed84-657f-4b26-82d9-5efa14541b33","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T20:29:27.984212Z","strongest_claim":"Experiments with state-of-the-art LLMs and LLM-based KGQA methods reveal their significant shortcomings, including frequent factual hallucinations and failures in causal reasoning.","one_line_summary":"CommonWhy is a new dataset of 15,000 why-questions for evaluating LLMs on entity-based causal commonsense reasoning grounded in Wikidata.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The questions in CommonWhy require genuine integration of entity facts with causal commonsense reasoning rather than being solvable through superficial patterns learned during training.","pith_extraction_headline":"CommonWhy introduces 15,000 why questions that test whether LLMs can combine specific entity facts with causal commonsense inference"},"references":{"count":70,"sample":[{"doi":"","year":2005,"title":"Satanjeev Banerjee and Alon Lavie. 2005. METEOR: An automatic metric for MT evaluation with improved correlation with human judgments. InProceedings of the acl workshop on intrinsic and extrinsic eval","work_id":"a32d6304-97ed-46fc-b35e-43f83cb5451e","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2017,"title":"CommAI: Evaluating the first steps towards a useful general AI","work_id":"e668a18c-b63a-4658-aaaf-8ee976ba8d0e","ref_index":2,"cited_arxiv_id":"1701.08954","is_internal_anchor":true},{"doi":"","year":2013,"title":"Jonathan Berant, Andrew Chou, Roy Frostig, and Percy Liang. 2013. Seman- tic parsing on freebase from question-answer pairs. InProceedings of the 2013 conference on empirical methods in natural langua","work_id":"ba594b62-da85-4be6-8345-1820b66064a3","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2013,"title":"Jonathan Berant, Andrew Chou, Roy Frostig, and Percy Liang. 2013. Semantic Parsing on Freebase from Question-Answer Pairs. InProceedings of the 2013 Conference on Empirical Methods in Natural Language","work_id":"a9a8461f-af0f-446a-bf88-c98eb912f05f","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"A is B” fail to learn “B is A","work_id":"e5438272-90bc-42ce-b08e-4f225459ecec","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":70,"snapshot_sha256":"daeb71380d12607fc77b9a974812d71deefc70c673faba9a67b39a96531e0e95","internal_anchors":3},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}