{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:TIBNXCRO6ESRTRGPN4UN2XZCIR","short_pith_number":"pith:TIBNXCRO","schema_version":"1.0","canonical_sha256":"9a02db8a2ef12519c4cf6f28dd5f22445a724e4c6caa37caac63f345619c214a","source":{"kind":"arxiv","id":"2507.11742","version":2},"attestation_state":"computed","paper":{"title":"CRABS: A syntactic-semantic pincer strategy for bounding LLM interpretation of Python notebooks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bertram Lud\\\"ascher, Dingmin Wang, Meng Li, Shin-Rong Tsai, Timothy M. McPhillips","submitted_at":"2025-07-15T21:14:08Z","abstract_excerpt":"Recognizing the information flows and operations comprising data science and machine learning Python notebooks is critical for evaluating, reusing, and adapting notebooks for new tasks. Investigating a notebook via re-execution often is impractical due to the challenges of resolving data and software dependencies. While Large Language Models (LLMs) pre-trained on large codebases have demonstrated effectiveness in understanding code without running it, we observe that they fail to understand some realistic notebooks due to hallucinations and long-context challenges. To address these issues, we "},"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":"2507.11742","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-15T21:14:08Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"e1e4776899d56f0d5245c4e10bf83b9cc9a980dfb22e33b2db015821dc792370","abstract_canon_sha256":"6c8959127d31b2a55f615f8c494bc483ab0057e927920f4c97953bd9c853069d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:58:32.791451Z","signature_b64":"M6+riteb9jXHFT+pVV8cJOdsuzFBE86zZjWiZgMwfQMxzV1Lam/2iyj7fWAlINIpniuTt5TbdPe4pL3ZlwHRBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9a02db8a2ef12519c4cf6f28dd5f22445a724e4c6caa37caac63f345619c214a","last_reissued_at":"2026-07-05T11:58:32.790910Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:58:32.790910Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CRABS: A syntactic-semantic pincer strategy for bounding LLM interpretation of Python notebooks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bertram Lud\\\"ascher, Dingmin Wang, Meng Li, Shin-Rong Tsai, Timothy M. McPhillips","submitted_at":"2025-07-15T21:14:08Z","abstract_excerpt":"Recognizing the information flows and operations comprising data science and machine learning Python notebooks is critical for evaluating, reusing, and adapting notebooks for new tasks. Investigating a notebook via re-execution often is impractical due to the challenges of resolving data and software dependencies. While Large Language Models (LLMs) pre-trained on large codebases have demonstrated effectiveness in understanding code without running it, we observe that they fail to understand some realistic notebooks due to hallucinations and long-context challenges. To address these issues, we "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.11742","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/2507.11742/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":"2507.11742","created_at":"2026-07-05T11:58:32.790968+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.11742v2","created_at":"2026-07-05T11:58:32.790968+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.11742","created_at":"2026-07-05T11:58:32.790968+00:00"},{"alias_kind":"pith_short_12","alias_value":"TIBNXCRO6ESR","created_at":"2026-07-05T11:58:32.790968+00:00"},{"alias_kind":"pith_short_16","alias_value":"TIBNXCRO6ESRTRGP","created_at":"2026-07-05T11:58:32.790968+00:00"},{"alias_kind":"pith_short_8","alias_value":"TIBNXCRO","created_at":"2026-07-05T11:58:32.790968+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.08687","citing_title":"PrepBench: How Far Are We from Natural-Language-Driven Data Preparation?","ref_index":32,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TIBNXCRO6ESRTRGPN4UN2XZCIR","json":"https://pith.science/pith/TIBNXCRO6ESRTRGPN4UN2XZCIR.json","graph_json":"https://pith.science/api/pith-number/TIBNXCRO6ESRTRGPN4UN2XZCIR/graph.json","events_json":"https://pith.science/api/pith-number/TIBNXCRO6ESRTRGPN4UN2XZCIR/events.json","paper":"https://pith.science/paper/TIBNXCRO"},"agent_actions":{"view_html":"https://pith.science/pith/TIBNXCRO6ESRTRGPN4UN2XZCIR","download_json":"https://pith.science/pith/TIBNXCRO6ESRTRGPN4UN2XZCIR.json","view_paper":"https://pith.science/paper/TIBNXCRO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.11742&json=true","fetch_graph":"https://pith.science/api/pith-number/TIBNXCRO6ESRTRGPN4UN2XZCIR/graph.json","fetch_events":"https://pith.science/api/pith-number/TIBNXCRO6ESRTRGPN4UN2XZCIR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TIBNXCRO6ESRTRGPN4UN2XZCIR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TIBNXCRO6ESRTRGPN4UN2XZCIR/action/storage_attestation","attest_author":"https://pith.science/pith/TIBNXCRO6ESRTRGPN4UN2XZCIR/action/author_attestation","sign_citation":"https://pith.science/pith/TIBNXCRO6ESRTRGPN4UN2XZCIR/action/citation_signature","submit_replication":"https://pith.science/pith/TIBNXCRO6ESRTRGPN4UN2XZCIR/action/replication_record"}},"created_at":"2026-07-05T11:58:32.790968+00:00","updated_at":"2026-07-05T11:58:32.790968+00:00"}