{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:E56CR3MBOJMNUIORPCLKPT5ROP","short_pith_number":"pith:E56CR3MB","schema_version":"1.0","canonical_sha256":"277c28ed817258da21d17896a7cfb173d6ec34357ac0a5d9d796d96e5c72ee87","source":{"kind":"arxiv","id":"2508.11260","version":1},"attestation_state":"computed","paper":{"title":"UNVEILING: What Makes Linguistics Olympiad Puzzles Tricky for LLMs?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Antara Raaghavi Bhattacharya, Dang Khoa Dang Dinh, Daria Kotova, Ekaterina Kochmar, Gaurja Aeron, Ikhlasul Akmal Hanif, KV Aditya Srivatsa, Monojit Choudhury, Mukund Choudhary","submitted_at":"2025-08-15T06:53:28Z","abstract_excerpt":"Large language models (LLMs) have demonstrated potential in reasoning tasks, but their performance on linguistics puzzles remains consistently poor. These puzzles, often derived from Linguistics Olympiad (LO) contests, provide a minimal contamination environment to assess LLMs' linguistic reasoning abilities across low-resource languages. This work analyses LLMs' performance on 629 problems across 41 low-resource languages by labelling each with linguistically informed features to unveil weaknesses. Our analyses show that LLMs struggle with puzzles involving higher morphological complexity and"},"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":"2508.11260","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-15T06:53:28Z","cross_cats_sorted":[],"title_canon_sha256":"e95af81686fc3ec6152fca22b3efa83ed010236b5e56b2006892477eb05b13f5","abstract_canon_sha256":"1dc9000f857f61320dfc2e695b5685d67bbf2b51a801e61ac0ba69aa0f72bfc8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:54:21.635635Z","signature_b64":"cptw6l7Fcw9igZBRW57cF+ZYP2RcitciVcEhAOWSU3CM66dHx6KPMZPFSkQ5Fy6/Ludx5vSc5x/IIfZ5RLlnBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"277c28ed817258da21d17896a7cfb173d6ec34357ac0a5d9d796d96e5c72ee87","last_reissued_at":"2026-07-05T11:54:21.635204Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:54:21.635204Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"UNVEILING: What Makes Linguistics Olympiad Puzzles Tricky for LLMs?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Antara Raaghavi Bhattacharya, Dang Khoa Dang Dinh, Daria Kotova, Ekaterina Kochmar, Gaurja Aeron, Ikhlasul Akmal Hanif, KV Aditya Srivatsa, Monojit Choudhury, Mukund Choudhary","submitted_at":"2025-08-15T06:53:28Z","abstract_excerpt":"Large language models (LLMs) have demonstrated potential in reasoning tasks, but their performance on linguistics puzzles remains consistently poor. These puzzles, often derived from Linguistics Olympiad (LO) contests, provide a minimal contamination environment to assess LLMs' linguistic reasoning abilities across low-resource languages. This work analyses LLMs' performance on 629 problems across 41 low-resource languages by labelling each with linguistically informed features to unveil weaknesses. Our analyses show that LLMs struggle with puzzles involving higher morphological complexity and"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.11260","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/2508.11260/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":"2508.11260","created_at":"2026-07-05T11:54:21.635271+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.11260v1","created_at":"2026-07-05T11:54:21.635271+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.11260","created_at":"2026-07-05T11:54:21.635271+00:00"},{"alias_kind":"pith_short_12","alias_value":"E56CR3MBOJMN","created_at":"2026-07-05T11:54:21.635271+00:00"},{"alias_kind":"pith_short_16","alias_value":"E56CR3MBOJMNUIOR","created_at":"2026-07-05T11:54:21.635271+00:00"},{"alias_kind":"pith_short_8","alias_value":"E56CR3MB","created_at":"2026-07-05T11:54:21.635271+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/E56CR3MBOJMNUIORPCLKPT5ROP","json":"https://pith.science/pith/E56CR3MBOJMNUIORPCLKPT5ROP.json","graph_json":"https://pith.science/api/pith-number/E56CR3MBOJMNUIORPCLKPT5ROP/graph.json","events_json":"https://pith.science/api/pith-number/E56CR3MBOJMNUIORPCLKPT5ROP/events.json","paper":"https://pith.science/paper/E56CR3MB"},"agent_actions":{"view_html":"https://pith.science/pith/E56CR3MBOJMNUIORPCLKPT5ROP","download_json":"https://pith.science/pith/E56CR3MBOJMNUIORPCLKPT5ROP.json","view_paper":"https://pith.science/paper/E56CR3MB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.11260&json=true","fetch_graph":"https://pith.science/api/pith-number/E56CR3MBOJMNUIORPCLKPT5ROP/graph.json","fetch_events":"https://pith.science/api/pith-number/E56CR3MBOJMNUIORPCLKPT5ROP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/E56CR3MBOJMNUIORPCLKPT5ROP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/E56CR3MBOJMNUIORPCLKPT5ROP/action/storage_attestation","attest_author":"https://pith.science/pith/E56CR3MBOJMNUIORPCLKPT5ROP/action/author_attestation","sign_citation":"https://pith.science/pith/E56CR3MBOJMNUIORPCLKPT5ROP/action/citation_signature","submit_replication":"https://pith.science/pith/E56CR3MBOJMNUIORPCLKPT5ROP/action/replication_record"}},"created_at":"2026-07-05T11:54:21.635271+00:00","updated_at":"2026-07-05T11:54:21.635271+00:00"}