{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:46O5BCYOANXWDGYJYAWJQFJLVK","short_pith_number":"pith:46O5BCYO","schema_version":"1.0","canonical_sha256":"e79dd08b0e036f619b09c02c98152baa8b9c97d7fb452ec2fd833d791b391de6","source":{"kind":"arxiv","id":"2411.07404","version":4},"attestation_state":"computed","paper":{"title":"Controllable Context Sensitivity and the Knob Behind It","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chris Wendler, Giovanni Monea, Julian Minder, Kevin Du, Niklas Stoehr, Robert West, Ryan Cotterell","submitted_at":"2024-11-11T22:22:21Z","abstract_excerpt":"When making predictions, a language model must trade off how much it relies on its context vs. its prior knowledge. Choosing how sensitive the model is to its context is a fundamental functionality, as it enables the model to excel at tasks like retrieval-augmented generation and question-answering. In this paper, we search for a knob which controls this sensitivity, determining whether language models answer from the context or their prior knowledge. To guide this search, we design a task for controllable context sensitivity. In this task, we first feed the model a context (Paris is in Englan"},"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":"2411.07404","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-11-11T22:22:21Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f5ddd7efb572ddc6c5107a5793dd0d2d5a173eda44f965d62760b9f5d6a16b18","abstract_canon_sha256":"e5ef96b340f737dcbd70379991def7065d590967563120d2629e444125fbf78c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:12:22.184983Z","signature_b64":"K0zKHM7o2Lyf4je6zTYdj1FxRhe5jT1Q8sCF41lJkRjSGEy0QZITRIzHY6XlKl0eLpNFsq0C/YkbGm5xTkExCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e79dd08b0e036f619b09c02c98152baa8b9c97d7fb452ec2fd833d791b391de6","last_reissued_at":"2026-07-05T11:12:22.184497Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:12:22.184497Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Controllable Context Sensitivity and the Knob Behind It","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chris Wendler, Giovanni Monea, Julian Minder, Kevin Du, Niklas Stoehr, Robert West, Ryan Cotterell","submitted_at":"2024-11-11T22:22:21Z","abstract_excerpt":"When making predictions, a language model must trade off how much it relies on its context vs. its prior knowledge. Choosing how sensitive the model is to its context is a fundamental functionality, as it enables the model to excel at tasks like retrieval-augmented generation and question-answering. In this paper, we search for a knob which controls this sensitivity, determining whether language models answer from the context or their prior knowledge. To guide this search, we design a task for controllable context sensitivity. In this task, we first feed the model a context (Paris is in Englan"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.07404","kind":"arxiv","version":4},"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/2411.07404/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":"2411.07404","created_at":"2026-07-05T11:12:22.184557+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.07404v4","created_at":"2026-07-05T11:12:22.184557+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.07404","created_at":"2026-07-05T11:12:22.184557+00:00"},{"alias_kind":"pith_short_12","alias_value":"46O5BCYOANXW","created_at":"2026-07-05T11:12:22.184557+00:00"},{"alias_kind":"pith_short_16","alias_value":"46O5BCYOANXWDGYJ","created_at":"2026-07-05T11:12:22.184557+00:00"},{"alias_kind":"pith_short_8","alias_value":"46O5BCYO","created_at":"2026-07-05T11:12:22.184557+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.19316","citing_title":"Sycophancy as compositions of Atomic Psychometric Traits","ref_index":21,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/46O5BCYOANXWDGYJYAWJQFJLVK","json":"https://pith.science/pith/46O5BCYOANXWDGYJYAWJQFJLVK.json","graph_json":"https://pith.science/api/pith-number/46O5BCYOANXWDGYJYAWJQFJLVK/graph.json","events_json":"https://pith.science/api/pith-number/46O5BCYOANXWDGYJYAWJQFJLVK/events.json","paper":"https://pith.science/paper/46O5BCYO"},"agent_actions":{"view_html":"https://pith.science/pith/46O5BCYOANXWDGYJYAWJQFJLVK","download_json":"https://pith.science/pith/46O5BCYOANXWDGYJYAWJQFJLVK.json","view_paper":"https://pith.science/paper/46O5BCYO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.07404&json=true","fetch_graph":"https://pith.science/api/pith-number/46O5BCYOANXWDGYJYAWJQFJLVK/graph.json","fetch_events":"https://pith.science/api/pith-number/46O5BCYOANXWDGYJYAWJQFJLVK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/46O5BCYOANXWDGYJYAWJQFJLVK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/46O5BCYOANXWDGYJYAWJQFJLVK/action/storage_attestation","attest_author":"https://pith.science/pith/46O5BCYOANXWDGYJYAWJQFJLVK/action/author_attestation","sign_citation":"https://pith.science/pith/46O5BCYOANXWDGYJYAWJQFJLVK/action/citation_signature","submit_replication":"https://pith.science/pith/46O5BCYOANXWDGYJYAWJQFJLVK/action/replication_record"}},"created_at":"2026-07-05T11:12:22.184557+00:00","updated_at":"2026-07-05T11:12:22.184557+00:00"}