{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:3EUXXFVUC6PVFG4BRSQ5H2QPN4","short_pith_number":"pith:3EUXXFVU","schema_version":"1.0","canonical_sha256":"d9297b96b4179f529b818ca1d3ea0f6f2786ddacf24adcefa21df547139576e8","source":{"kind":"arxiv","id":"2404.06283","version":2},"attestation_state":"computed","paper":{"title":"LLMs' Reading Comprehension Is Affected by Parametric Knowledge and Struggles with Hypothetical Statements","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Reut Tsarfaty, Victoria Basmov, Yoav Goldberg","submitted_at":"2024-04-09T13:08:56Z","abstract_excerpt":"The task of reading comprehension (RC), often implemented as context-based question answering (QA), provides a primary means to assess language models' natural language understanding (NLU) capabilities. Yet, when applied to large language models (LLMs) with extensive built-in world knowledge, this method can be deceptive. If the context aligns with the LLMs' internal knowledge, it is hard to discern whether the models' answers stem from context comprehension or from LLMs' internal information. Conversely, using data that conflicts with the models' knowledge creates erroneous trends which disto"},"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":"2404.06283","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-09T13:08:56Z","cross_cats_sorted":[],"title_canon_sha256":"63b0c182c037fdd9a6da682f8e93e0b01286c6c539ad13e8cdb505e9db5010b9","abstract_canon_sha256":"c5b1f40aa92813e5b76b4ca1fd82eb98c02aa6b102c61f108c7503b663f429ac"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:32:14.973931Z","signature_b64":"mlWnGDhPtVl91+kzO3Jufc7yOCpicBiiUKu2e1FitezlPnR5r4LjltKnL2hiClUkCIaZTeLJEw4NqUNL/LuyBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d9297b96b4179f529b818ca1d3ea0f6f2786ddacf24adcefa21df547139576e8","last_reissued_at":"2026-07-05T11:32:14.973440Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:32:14.973440Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LLMs' Reading Comprehension Is Affected by Parametric Knowledge and Struggles with Hypothetical Statements","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Reut Tsarfaty, Victoria Basmov, Yoav Goldberg","submitted_at":"2024-04-09T13:08:56Z","abstract_excerpt":"The task of reading comprehension (RC), often implemented as context-based question answering (QA), provides a primary means to assess language models' natural language understanding (NLU) capabilities. Yet, when applied to large language models (LLMs) with extensive built-in world knowledge, this method can be deceptive. If the context aligns with the LLMs' internal knowledge, it is hard to discern whether the models' answers stem from context comprehension or from LLMs' internal information. Conversely, using data that conflicts with the models' knowledge creates erroneous trends which disto"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.06283","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/2404.06283/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":"2404.06283","created_at":"2026-07-05T11:32:14.973492+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.06283v2","created_at":"2026-07-05T11:32:14.973492+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.06283","created_at":"2026-07-05T11:32:14.973492+00:00"},{"alias_kind":"pith_short_12","alias_value":"3EUXXFVUC6PV","created_at":"2026-07-05T11:32:14.973492+00:00"},{"alias_kind":"pith_short_16","alias_value":"3EUXXFVUC6PVFG4B","created_at":"2026-07-05T11:32:14.973492+00:00"},{"alias_kind":"pith_short_8","alias_value":"3EUXXFVU","created_at":"2026-07-05T11:32:14.973492+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.23790","citing_title":"Rethinking the Understanding Ability across LLMs through Mutual Information","ref_index":3,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3EUXXFVUC6PVFG4BRSQ5H2QPN4","json":"https://pith.science/pith/3EUXXFVUC6PVFG4BRSQ5H2QPN4.json","graph_json":"https://pith.science/api/pith-number/3EUXXFVUC6PVFG4BRSQ5H2QPN4/graph.json","events_json":"https://pith.science/api/pith-number/3EUXXFVUC6PVFG4BRSQ5H2QPN4/events.json","paper":"https://pith.science/paper/3EUXXFVU"},"agent_actions":{"view_html":"https://pith.science/pith/3EUXXFVUC6PVFG4BRSQ5H2QPN4","download_json":"https://pith.science/pith/3EUXXFVUC6PVFG4BRSQ5H2QPN4.json","view_paper":"https://pith.science/paper/3EUXXFVU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.06283&json=true","fetch_graph":"https://pith.science/api/pith-number/3EUXXFVUC6PVFG4BRSQ5H2QPN4/graph.json","fetch_events":"https://pith.science/api/pith-number/3EUXXFVUC6PVFG4BRSQ5H2QPN4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3EUXXFVUC6PVFG4BRSQ5H2QPN4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3EUXXFVUC6PVFG4BRSQ5H2QPN4/action/storage_attestation","attest_author":"https://pith.science/pith/3EUXXFVUC6PVFG4BRSQ5H2QPN4/action/author_attestation","sign_citation":"https://pith.science/pith/3EUXXFVUC6PVFG4BRSQ5H2QPN4/action/citation_signature","submit_replication":"https://pith.science/pith/3EUXXFVUC6PVFG4BRSQ5H2QPN4/action/replication_record"}},"created_at":"2026-07-05T11:32:14.973492+00:00","updated_at":"2026-07-05T11:32:14.973492+00:00"}