{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:QATNLBGKX4LURJVT6HUG4DFRTO","short_pith_number":"pith:QATNLBGK","canonical_record":{"source":{"id":"2412.09012","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-12T07:23:52Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"674c109d8ed9bbeb26b9c577cf41d65af70d50d9b18ef6fd174ccc3763243d7f","abstract_canon_sha256":"3e63f1229f6e96414ee8c29acfa42719bf497d40ca05d7fefd638d6c98e3d013"},"schema_version":"1.0"},"canonical_sha256":"8026d584cabf1748a6b3f1e86e0cb19b973fa7b80b9723d6b7335f8cc7cad849","source":{"kind":"arxiv","id":"2412.09012","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.09012","created_at":"2026-07-05T10:00:44Z"},{"alias_kind":"arxiv_version","alias_value":"2412.09012v2","created_at":"2026-07-05T10:00:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.09012","created_at":"2026-07-05T10:00:44Z"},{"alias_kind":"pith_short_12","alias_value":"QATNLBGKX4LU","created_at":"2026-07-05T10:00:44Z"},{"alias_kind":"pith_short_16","alias_value":"QATNLBGKX4LURJVT","created_at":"2026-07-05T10:00:44Z"},{"alias_kind":"pith_short_8","alias_value":"QATNLBGK","created_at":"2026-07-05T10:00:44Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:QATNLBGKX4LURJVT6HUG4DFRTO","target":"record","payload":{"canonical_record":{"source":{"id":"2412.09012","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-12T07:23:52Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"674c109d8ed9bbeb26b9c577cf41d65af70d50d9b18ef6fd174ccc3763243d7f","abstract_canon_sha256":"3e63f1229f6e96414ee8c29acfa42719bf497d40ca05d7fefd638d6c98e3d013"},"schema_version":"1.0"},"canonical_sha256":"8026d584cabf1748a6b3f1e86e0cb19b973fa7b80b9723d6b7335f8cc7cad849","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:00:44.946461Z","signature_b64":"Scvbwz61lt+poHyBmbEh5V6pznlgA5hCxtRVxymbuSziUpODzGpdjxKSEI4OWnom6PlcPOXN5iigSK90YxXXAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8026d584cabf1748a6b3f1e86e0cb19b973fa7b80b9723d6b7335f8cc7cad849","last_reissued_at":"2026-07-05T10:00:44.945960Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:00:44.945960Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2412.09012","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:00:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"b3ObknBSYY609yLmTmsisEygBKAXPArbcEfihOtyzxuYr6IS8QZByvE3yS5UnyidAWVyxCuYW1+77/UnaapgAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T06:40:23.985413Z"},"content_sha256":"3efa959a08105c8c5c13312022e88c0e1655e0c82f7ba8304a58c982cf81d9de","schema_version":"1.0","event_id":"sha256:3efa959a08105c8c5c13312022e88c0e1655e0c82f7ba8304a58c982cf81d9de"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:QATNLBGKX4LURJVT6HUG4DFRTO","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"What Makes Cryptic Crosswords Challenging for LLMs?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Abdelrahman Sadallah, Daria Kotova, Ekaterina Kochmar","submitted_at":"2024-12-12T07:23:52Z","abstract_excerpt":"Cryptic crosswords are puzzles that rely on general knowledge and the solver's ability to manipulate language on different levels, dealing with various types of wordplay. Previous research suggests that solving such puzzles is challenging even for modern NLP models, including Large Language Models (LLMs). However, there is little to no research on the reasons for their poor performance on this task. In this paper, we establish the benchmark results for three popular LLMs: Gemma2, LLaMA3 and ChatGPT, showing that their performance on this task is still significantly below that of humans. We als"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.09012","kind":"arxiv","version":2},"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/2412.09012/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:00:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6zR79cMT4tFt3RKccRehELRRLq06GB44V4wb/l/CItcUCi45uiAkUq5kNltARzD9wGY6+/CIEDcV0VfysJ1nDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T06:40:23.985902Z"},"content_sha256":"3c7634479b18febea6ae273386f1066d68dbf59a22c33e93ad6d08554b525f1b","schema_version":"1.0","event_id":"sha256:3c7634479b18febea6ae273386f1066d68dbf59a22c33e93ad6d08554b525f1b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QATNLBGKX4LURJVT6HUG4DFRTO/bundle.json","state_url":"https://pith.science/pith/QATNLBGKX4LURJVT6HUG4DFRTO/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QATNLBGKX4LURJVT6HUG4DFRTO/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-14T06:40:23Z","links":{"resolver":"https://pith.science/pith/QATNLBGKX4LURJVT6HUG4DFRTO","bundle":"https://pith.science/pith/QATNLBGKX4LURJVT6HUG4DFRTO/bundle.json","state":"https://pith.science/pith/QATNLBGKX4LURJVT6HUG4DFRTO/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QATNLBGKX4LURJVT6HUG4DFRTO/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:QATNLBGKX4LURJVT6HUG4DFRTO","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":"3e63f1229f6e96414ee8c29acfa42719bf497d40ca05d7fefd638d6c98e3d013","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-12T07:23:52Z","title_canon_sha256":"674c109d8ed9bbeb26b9c577cf41d65af70d50d9b18ef6fd174ccc3763243d7f"},"schema_version":"1.0","source":{"id":"2412.09012","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.09012","created_at":"2026-07-05T10:00:44Z"},{"alias_kind":"arxiv_version","alias_value":"2412.09012v2","created_at":"2026-07-05T10:00:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.09012","created_at":"2026-07-05T10:00:44Z"},{"alias_kind":"pith_short_12","alias_value":"QATNLBGKX4LU","created_at":"2026-07-05T10:00:44Z"},{"alias_kind":"pith_short_16","alias_value":"QATNLBGKX4LURJVT","created_at":"2026-07-05T10:00:44Z"},{"alias_kind":"pith_short_8","alias_value":"QATNLBGK","created_at":"2026-07-05T10:00:44Z"}],"graph_snapshots":[{"event_id":"sha256:3c7634479b18febea6ae273386f1066d68dbf59a22c33e93ad6d08554b525f1b","target":"graph","created_at":"2026-07-05T10:00:44Z","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/2412.09012/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Cryptic crosswords are puzzles that rely on general knowledge and the solver's ability to manipulate language on different levels, dealing with various types of wordplay. Previous research suggests that solving such puzzles is challenging even for modern NLP models, including Large Language Models (LLMs). However, there is little to no research on the reasons for their poor performance on this task. In this paper, we establish the benchmark results for three popular LLMs: Gemma2, LLaMA3 and ChatGPT, showing that their performance on this task is still significantly below that of humans. We als","authors_text":"Abdelrahman Sadallah, Daria Kotova, Ekaterina Kochmar","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-12T07:23:52Z","title":"What Makes Cryptic Crosswords Challenging for LLMs?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.09012","kind":"arxiv","version":2},"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":"sha256:3efa959a08105c8c5c13312022e88c0e1655e0c82f7ba8304a58c982cf81d9de","target":"record","created_at":"2026-07-05T10:00:44Z","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":"3e63f1229f6e96414ee8c29acfa42719bf497d40ca05d7fefd638d6c98e3d013","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-12T07:23:52Z","title_canon_sha256":"674c109d8ed9bbeb26b9c577cf41d65af70d50d9b18ef6fd174ccc3763243d7f"},"schema_version":"1.0","source":{"id":"2412.09012","kind":"arxiv","version":2}},"canonical_sha256":"8026d584cabf1748a6b3f1e86e0cb19b973fa7b80b9723d6b7335f8cc7cad849","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8026d584cabf1748a6b3f1e86e0cb19b973fa7b80b9723d6b7335f8cc7cad849","first_computed_at":"2026-07-05T10:00:44.945960Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:00:44.945960Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Scvbwz61lt+poHyBmbEh5V6pznlgA5hCxtRVxymbuSziUpODzGpdjxKSEI4OWnom6PlcPOXN5iigSK90YxXXAw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:00:44.946461Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.09012","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3efa959a08105c8c5c13312022e88c0e1655e0c82f7ba8304a58c982cf81d9de","sha256:3c7634479b18febea6ae273386f1066d68dbf59a22c33e93ad6d08554b525f1b"],"state_sha256":"729c79870799ebda8a7f28f92694a8aa8def94a7fe40634840c2704b1a8d4df6"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5bgrvRVW/ROP+PPkDckxrfoHs3jK/QSA9WKle7pTCI9HU1xavGXbFs3Nv+wA2viAyR9bYaSOdEUcLr1ONSj9Bg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-14T06:40:23.991056Z","bundle_sha256":"43a8f5071b371718f09312f30d88b391f931cf5bb34170646b726c84a1fade9e"}}