{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:2YSAYIPVSCKKJP277SESYLAUNG","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"54962475479f1f4a735ccb247004d197d061c6b3e0518483590a162a26457906","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-20T02:47:18Z","title_canon_sha256":"f3ac53b7ddc9fb5ee973490294c4cbca8c947eaecd805e4cbc7b9d2e8eba014a"},"schema_version":"1.0","source":{"id":"2504.14462","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.14462","created_at":"2026-07-05T10:51:43Z"},{"alias_kind":"arxiv_version","alias_value":"2504.14462v1","created_at":"2026-07-05T10:51:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.14462","created_at":"2026-07-05T10:51:43Z"},{"alias_kind":"pith_short_12","alias_value":"2YSAYIPVSCKK","created_at":"2026-07-05T10:51:43Z"},{"alias_kind":"pith_short_16","alias_value":"2YSAYIPVSCKKJP27","created_at":"2026-07-05T10:51:43Z"},{"alias_kind":"pith_short_8","alias_value":"2YSAYIPV","created_at":"2026-07-05T10:51:43Z"}],"graph_snapshots":[{"event_id":"sha256:141e95ff87d28e8ddf60f7d62af339a74c2bec5ada782211bcec382cf4d6ba29","target":"graph","created_at":"2026-07-05T10:51:43Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2504.14462/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The rise of Large Language Models (LLMs) has redefined the AI landscape, particularly due to their ability to encode factual and commonsense knowledge, and their outstanding performance in tasks requiring reasoning. Despite these advances, hallucinations and reasoning errors remain a significant barrier to their deployment in high-stakes settings. In this work, we observe that even the most prominent LLMs, such as OpenAI-o1, suffer from high rates of reasoning errors and hallucinations on tasks requiring commonsense reasoning over obscure, long-tail entities. To investigate this limitation, we","authors_text":"Armin Toroghi, Scott Sanner, Willis Guo","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-20T02:47:18Z","title":"CoLoTa: A Dataset for Entity-based Commonsense Reasoning over Long-Tail Knowledge"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.14462","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:2c42e07ccd74b640a9ae161145e835ac565583bb2167c7ab465b846eb5b2d97e","target":"record","created_at":"2026-07-05T10:51:43Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"54962475479f1f4a735ccb247004d197d061c6b3e0518483590a162a26457906","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-20T02:47:18Z","title_canon_sha256":"f3ac53b7ddc9fb5ee973490294c4cbca8c947eaecd805e4cbc7b9d2e8eba014a"},"schema_version":"1.0","source":{"id":"2504.14462","kind":"arxiv","version":1}},"canonical_sha256":"d6240c21f59094a4bf5ffc892c2c1469abed0569d0d679ed2aa24bc9c1689f9b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d6240c21f59094a4bf5ffc892c2c1469abed0569d0d679ed2aa24bc9c1689f9b","first_computed_at":"2026-07-05T10:51:43.764987Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:51:43.764987Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"790mIlBu/AZQiqrOzfaqh7BeNPDzcaUDNAL0yvZve9RD02M2oPuMRL10PnsSTeEzuHDGyz28X63iwJTRSJsTCA==","signature_status":"signed_v1","signed_at":"2026-07-05T10:51:43.765647Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.14462","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2c42e07ccd74b640a9ae161145e835ac565583bb2167c7ab465b846eb5b2d97e","sha256:141e95ff87d28e8ddf60f7d62af339a74c2bec5ada782211bcec382cf4d6ba29"],"state_sha256":"5395728c0e5ef0054ecd82cf9c010ca303d8ce2da03fe290ff2ccbcf4b1a626d"}