{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:TBOGOJTN7CBRMCHBYJ2BKXSX3M","short_pith_number":"pith:TBOGOJTN","schema_version":"1.0","canonical_sha256":"985c67266df8831608e1c274155e57db032596b23dc90a565187686352301bd0","source":{"kind":"arxiv","id":"2311.15930","version":1},"attestation_state":"computed","paper":{"title":"WorldSense: A Synthetic Benchmark for Grounded Reasoning in Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Dieuwke Hupkes, Emmanuel Dupoux, Gr\\'egoire Mialon, Jean-Baptiste Gaya, Mark Ibrahim, Megi Dervishi, Pascal Vincent, Thomas Scialom, Xavier Martinet, Youssef Benchekroun","submitted_at":"2023-11-27T15:38:17Z","abstract_excerpt":"We propose WorldSense, a benchmark designed to assess the extent to which LLMs are consistently able to sustain tacit world models, by testing how they draw simple inferences from descriptions of simple arrangements of entities. Worldsense is a synthetic benchmark with three problem types, each with their own trivial control, which explicitly avoids bias by decorrelating the abstract structure of problems from the vocabulary and expressions, and by decorrelating all problem subparts with the correct response. We run our benchmark on three state-of-the-art chat-LLMs (GPT3.5, GPT4 and Llama2-cha"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2311.15930","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-11-27T15:38:17Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9e3c4ca0f67c130ba2894245601535e1cbf14f29c691ca2620ff448e0187950f","abstract_canon_sha256":"e9cdd00e6e6adb9421fa2d4d24dfc9891014d4add7bfafa7104d781a0bf3434e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:17:11.395609Z","signature_b64":"n5nHb30q3xwIikf4otdUhZOelQLk88DlwmS0Zh5INtJz/1qGOTEE0hhbcnQpyof/wmMCtKlGyxxdhatlCtepAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"985c67266df8831608e1c274155e57db032596b23dc90a565187686352301bd0","last_reissued_at":"2026-07-05T07:17:11.395127Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:17:11.395127Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"WorldSense: A Synthetic Benchmark for Grounded Reasoning in Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Dieuwke Hupkes, Emmanuel Dupoux, Gr\\'egoire Mialon, Jean-Baptiste Gaya, Mark Ibrahim, Megi Dervishi, Pascal Vincent, Thomas Scialom, Xavier Martinet, Youssef Benchekroun","submitted_at":"2023-11-27T15:38:17Z","abstract_excerpt":"We propose WorldSense, a benchmark designed to assess the extent to which LLMs are consistently able to sustain tacit world models, by testing how they draw simple inferences from descriptions of simple arrangements of entities. Worldsense is a synthetic benchmark with three problem types, each with their own trivial control, which explicitly avoids bias by decorrelating the abstract structure of problems from the vocabulary and expressions, and by decorrelating all problem subparts with the correct response. We run our benchmark on three state-of-the-art chat-LLMs (GPT3.5, GPT4 and Llama2-cha"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.15930","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2311.15930/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2311.15930","created_at":"2026-07-05T07:17:11.395182+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.15930v1","created_at":"2026-07-05T07:17:11.395182+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.15930","created_at":"2026-07-05T07:17:11.395182+00:00"},{"alias_kind":"pith_short_12","alias_value":"TBOGOJTN7CBR","created_at":"2026-07-05T07:17:11.395182+00:00"},{"alias_kind":"pith_short_16","alias_value":"TBOGOJTN7CBRMCHB","created_at":"2026-07-05T07:17:11.395182+00:00"},{"alias_kind":"pith_short_8","alias_value":"TBOGOJTN","created_at":"2026-07-05T07:17:11.395182+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.28733","citing_title":"Agentic Abstention: Do Agents Know When to Stop Instead of Act?","ref_index":79,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00869","citing_title":"Enhancing LLM Metacognition via Cognitive Pairwise Training","ref_index":75,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TBOGOJTN7CBRMCHBYJ2BKXSX3M","json":"https://pith.science/pith/TBOGOJTN7CBRMCHBYJ2BKXSX3M.json","graph_json":"https://pith.science/api/pith-number/TBOGOJTN7CBRMCHBYJ2BKXSX3M/graph.json","events_json":"https://pith.science/api/pith-number/TBOGOJTN7CBRMCHBYJ2BKXSX3M/events.json","paper":"https://pith.science/paper/TBOGOJTN"},"agent_actions":{"view_html":"https://pith.science/pith/TBOGOJTN7CBRMCHBYJ2BKXSX3M","download_json":"https://pith.science/pith/TBOGOJTN7CBRMCHBYJ2BKXSX3M.json","view_paper":"https://pith.science/paper/TBOGOJTN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.15930&json=true","fetch_graph":"https://pith.science/api/pith-number/TBOGOJTN7CBRMCHBYJ2BKXSX3M/graph.json","fetch_events":"https://pith.science/api/pith-number/TBOGOJTN7CBRMCHBYJ2BKXSX3M/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TBOGOJTN7CBRMCHBYJ2BKXSX3M/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TBOGOJTN7CBRMCHBYJ2BKXSX3M/action/storage_attestation","attest_author":"https://pith.science/pith/TBOGOJTN7CBRMCHBYJ2BKXSX3M/action/author_attestation","sign_citation":"https://pith.science/pith/TBOGOJTN7CBRMCHBYJ2BKXSX3M/action/citation_signature","submit_replication":"https://pith.science/pith/TBOGOJTN7CBRMCHBYJ2BKXSX3M/action/replication_record"}},"created_at":"2026-07-05T07:17:11.395182+00:00","updated_at":"2026-07-05T07:17:11.395182+00:00"}