{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:NWLC5F2SCMWUM5V435CEIJQH5M","short_pith_number":"pith:NWLC5F2S","schema_version":"1.0","canonical_sha256":"6d962e9752132d4676bcdf44442607eb3ea8283e540db4a6df9df7c03ba40034","source":{"kind":"arxiv","id":"2301.11916","version":4},"attestation_state":"computed","paper":{"title":"Large Language Models Are Latent Variable Models: Explaining and Finding Good Demonstrations for In-Context Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Mark Steyvers, Michael Saxon, Wanrong Zhu, William Yang Wang, Xinyi Wang","submitted_at":"2023-01-27T18:59:01Z","abstract_excerpt":"In recent years, pre-trained large language models (LLMs) have demonstrated remarkable efficiency in achieving an inference-time few-shot learning capability known as in-context learning. However, existing literature has highlighted the sensitivity of this capability to the selection of few-shot demonstrations. Current understandings of the underlying mechanisms by which this capability arises from regular language model pretraining objectives remain disconnected from the real-world LLMs. This study aims to examine the in-context learning phenomenon through a Bayesian lens, viewing real-world "},"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":"2301.11916","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-01-27T18:59:01Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"960a28a149b5024a36fcb48f08124f2f175109d35bc34ce5b077e1e88b838887","abstract_canon_sha256":"fd07b14c65621cc4766dfe7ec6ffd2af5c727706663530faf32cf43e73afb9f8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:44:16.604725Z","signature_b64":"0wGmJumwkBqyZp9a11TOmI62LcPtZnq/lJWX8ck3eXAQNfh2G5uXuhcdt76UaY1mGy9lDeqK8eY3dN0b5U+gDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6d962e9752132d4676bcdf44442607eb3ea8283e540db4a6df9df7c03ba40034","last_reissued_at":"2026-07-05T07:44:16.604234Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:44:16.604234Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Large Language Models Are Latent Variable Models: Explaining and Finding Good Demonstrations for In-Context Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Mark Steyvers, Michael Saxon, Wanrong Zhu, William Yang Wang, Xinyi Wang","submitted_at":"2023-01-27T18:59:01Z","abstract_excerpt":"In recent years, pre-trained large language models (LLMs) have demonstrated remarkable efficiency in achieving an inference-time few-shot learning capability known as in-context learning. However, existing literature has highlighted the sensitivity of this capability to the selection of few-shot demonstrations. Current understandings of the underlying mechanisms by which this capability arises from regular language model pretraining objectives remain disconnected from the real-world LLMs. This study aims to examine the in-context learning phenomenon through a Bayesian lens, viewing real-world "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2301.11916","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/2301.11916/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":"2301.11916","created_at":"2026-07-05T07:44:16.604295+00:00"},{"alias_kind":"arxiv_version","alias_value":"2301.11916v4","created_at":"2026-07-05T07:44:16.604295+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2301.11916","created_at":"2026-07-05T07:44:16.604295+00:00"},{"alias_kind":"pith_short_12","alias_value":"NWLC5F2SCMWU","created_at":"2026-07-05T07:44:16.604295+00:00"},{"alias_kind":"pith_short_16","alias_value":"NWLC5F2SCMWUM5V4","created_at":"2026-07-05T07:44:16.604295+00:00"},{"alias_kind":"pith_short_8","alias_value":"NWLC5F2S","created_at":"2026-07-05T07:44:16.604295+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.00545","citing_title":"The Assistant as a Privileged Persona: A canonical reference in cross-persona self-recognition","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2506.07298","citing_title":"Pre-trained Large Language Models Learn Hidden Markov Models In-context","ref_index":54,"is_internal_anchor":false},{"citing_arxiv_id":"2510.18117","citing_title":"Online In-Context Distillation for Low-Resource Vision Language Models","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09727","citing_title":"One for All: A Non-Linear Transformer can Enable Cross-Domain Generalization for In-Context Reinforcement Learning","ref_index":20,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NWLC5F2SCMWUM5V435CEIJQH5M","json":"https://pith.science/pith/NWLC5F2SCMWUM5V435CEIJQH5M.json","graph_json":"https://pith.science/api/pith-number/NWLC5F2SCMWUM5V435CEIJQH5M/graph.json","events_json":"https://pith.science/api/pith-number/NWLC5F2SCMWUM5V435CEIJQH5M/events.json","paper":"https://pith.science/paper/NWLC5F2S"},"agent_actions":{"view_html":"https://pith.science/pith/NWLC5F2SCMWUM5V435CEIJQH5M","download_json":"https://pith.science/pith/NWLC5F2SCMWUM5V435CEIJQH5M.json","view_paper":"https://pith.science/paper/NWLC5F2S","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2301.11916&json=true","fetch_graph":"https://pith.science/api/pith-number/NWLC5F2SCMWUM5V435CEIJQH5M/graph.json","fetch_events":"https://pith.science/api/pith-number/NWLC5F2SCMWUM5V435CEIJQH5M/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NWLC5F2SCMWUM5V435CEIJQH5M/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NWLC5F2SCMWUM5V435CEIJQH5M/action/storage_attestation","attest_author":"https://pith.science/pith/NWLC5F2SCMWUM5V435CEIJQH5M/action/author_attestation","sign_citation":"https://pith.science/pith/NWLC5F2SCMWUM5V435CEIJQH5M/action/citation_signature","submit_replication":"https://pith.science/pith/NWLC5F2SCMWUM5V435CEIJQH5M/action/replication_record"}},"created_at":"2026-07-05T07:44:16.604295+00:00","updated_at":"2026-07-05T07:44:16.604295+00:00"}