{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:WBROMWWK4WGMG2AMC7JXEBWZ2T","short_pith_number":"pith:WBROMWWK","schema_version":"1.0","canonical_sha256":"b062e65acae58cc3680c17d37206d9d4d4b66b1e8ee36fe99249570cfadfc36e","source":{"kind":"arxiv","id":"2412.03782","version":3},"attestation_state":"computed","paper":{"title":"The broader spectrum of in-context learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Aaditya K. Singh, Andrew Kyle Lampinen, Murray Shanahan, Stephanie C. Y. Chan","submitted_at":"2024-12-05T00:05:11Z","abstract_excerpt":"The ability of language models to learn a task from a few examples in context has generated substantial interest. Here, we provide a perspective that situates this type of supervised few-shot learning within a much broader spectrum of meta-learned in-context learning. Indeed, we suggest that any distribution of sequences in which context non-trivially decreases loss on subsequent predictions can be interpreted as eliciting a kind of in-context learning. We suggest that this perspective helps to unify the broad set of in-context abilities that language models exhibit -- such as adapting to task"},"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":"2412.03782","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-05T00:05:11Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"f46e67331538937ec12b2d1d70da08cb35c20622d73a71172bfb445d413831e7","abstract_canon_sha256":"10dcfd9ec8fc7cd251e89f7cb909f7a0b708901dbbb762febd59187a695683c5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:16:30.975130Z","signature_b64":"WlZ7njXxMU2f7oKqVuJcpcJWm4pAQgxL8tah33dS/GOaKR7+3vX5NQcHy5eP03qadKsMxq2Ng5g4Twxw9XwbBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b062e65acae58cc3680c17d37206d9d4d4b66b1e8ee36fe99249570cfadfc36e","last_reissued_at":"2026-07-05T11:16:30.974604Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:16:30.974604Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The broader spectrum of in-context learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Aaditya K. Singh, Andrew Kyle Lampinen, Murray Shanahan, Stephanie C. Y. Chan","submitted_at":"2024-12-05T00:05:11Z","abstract_excerpt":"The ability of language models to learn a task from a few examples in context has generated substantial interest. Here, we provide a perspective that situates this type of supervised few-shot learning within a much broader spectrum of meta-learned in-context learning. Indeed, we suggest that any distribution of sequences in which context non-trivially decreases loss on subsequent predictions can be interpreted as eliciting a kind of in-context learning. We suggest that this perspective helps to unify the broad set of in-context abilities that language models exhibit -- such as adapting to task"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.03782","kind":"arxiv","version":3},"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/2412.03782/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":"2412.03782","created_at":"2026-07-05T11:16:30.974669+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.03782v3","created_at":"2026-07-05T11:16:30.974669+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.03782","created_at":"2026-07-05T11:16:30.974669+00:00"},{"alias_kind":"pith_short_12","alias_value":"WBROMWWK4WGM","created_at":"2026-07-05T11:16:30.974669+00:00"},{"alias_kind":"pith_short_16","alias_value":"WBROMWWK4WGMG2AM","created_at":"2026-07-05T11:16:30.974669+00:00"},{"alias_kind":"pith_short_8","alias_value":"WBROMWWK","created_at":"2026-07-05T11:16:30.974669+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2602.07794","citing_title":"Emergent Structured Representations Support Flexible In-Context Inference in Large Language Models","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2509.20328","citing_title":"Video models are zero-shot learners and reasoners","ref_index":69,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12412","citing_title":"Stories in Space: In-Context Learning Trajectories in Conceptual Belief Space","ref_index":118,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05115","citing_title":"Manifold Steering Reveals the Shared Geometry of Neural Network Representation and Behavior","ref_index":254,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WBROMWWK4WGMG2AMC7JXEBWZ2T","json":"https://pith.science/pith/WBROMWWK4WGMG2AMC7JXEBWZ2T.json","graph_json":"https://pith.science/api/pith-number/WBROMWWK4WGMG2AMC7JXEBWZ2T/graph.json","events_json":"https://pith.science/api/pith-number/WBROMWWK4WGMG2AMC7JXEBWZ2T/events.json","paper":"https://pith.science/paper/WBROMWWK"},"agent_actions":{"view_html":"https://pith.science/pith/WBROMWWK4WGMG2AMC7JXEBWZ2T","download_json":"https://pith.science/pith/WBROMWWK4WGMG2AMC7JXEBWZ2T.json","view_paper":"https://pith.science/paper/WBROMWWK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.03782&json=true","fetch_graph":"https://pith.science/api/pith-number/WBROMWWK4WGMG2AMC7JXEBWZ2T/graph.json","fetch_events":"https://pith.science/api/pith-number/WBROMWWK4WGMG2AMC7JXEBWZ2T/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WBROMWWK4WGMG2AMC7JXEBWZ2T/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WBROMWWK4WGMG2AMC7JXEBWZ2T/action/storage_attestation","attest_author":"https://pith.science/pith/WBROMWWK4WGMG2AMC7JXEBWZ2T/action/author_attestation","sign_citation":"https://pith.science/pith/WBROMWWK4WGMG2AMC7JXEBWZ2T/action/citation_signature","submit_replication":"https://pith.science/pith/WBROMWWK4WGMG2AMC7JXEBWZ2T/action/replication_record"}},"created_at":"2026-07-05T11:16:30.974669+00:00","updated_at":"2026-07-05T11:16:30.974669+00:00"}