{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:VPSR4ZICWER5DNM3ZWX4JTRFRJ","short_pith_number":"pith:VPSR4ZIC","canonical_record":{"source":{"id":"2111.02080","kind":"arxiv","version":6},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-11-03T09:12:33Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"d11121fff8b4527ebb3daba0fca9697ace21e7716128ef726203c70bc0afe3a2","abstract_canon_sha256":"ece93fb738dd7f046125c81a57fa34c43199dcc6730005f0994bac4c271d5308"},"schema_version":"1.0"},"canonical_sha256":"abe51e6502b123d1b59bcdafc4ce258a748628095a8fba3c262f049ff320956a","source":{"kind":"arxiv","id":"2111.02080","version":6},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2111.02080","created_at":"2026-05-17T23:38:46Z"},{"alias_kind":"arxiv_version","alias_value":"2111.02080v6","created_at":"2026-05-17T23:38:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2111.02080","created_at":"2026-05-17T23:38:46Z"},{"alias_kind":"pith_short_12","alias_value":"VPSR4ZICWER5","created_at":"2026-05-18T12:33:33Z"},{"alias_kind":"pith_short_16","alias_value":"VPSR4ZICWER5DNM3","created_at":"2026-05-18T12:33:33Z"},{"alias_kind":"pith_short_8","alias_value":"VPSR4ZIC","created_at":"2026-05-18T12:33:33Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:VPSR4ZICWER5DNM3ZWX4JTRFRJ","target":"record","payload":{"canonical_record":{"source":{"id":"2111.02080","kind":"arxiv","version":6},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-11-03T09:12:33Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"d11121fff8b4527ebb3daba0fca9697ace21e7716128ef726203c70bc0afe3a2","abstract_canon_sha256":"ece93fb738dd7f046125c81a57fa34c43199dcc6730005f0994bac4c271d5308"},"schema_version":"1.0"},"canonical_sha256":"abe51e6502b123d1b59bcdafc4ce258a748628095a8fba3c262f049ff320956a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:38:46.404663Z","signature_b64":"OrzOJeld4FxePDZPLH4TxbHTvtLJLKQLSIoJs8dF+mk+p42dqTFl0PLpky+KLEdZxnTt+4kSc7y+QDGlGKiZCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"abe51e6502b123d1b59bcdafc4ce258a748628095a8fba3c262f049ff320956a","last_reissued_at":"2026-05-17T23:38:46.404101Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:38:46.404101Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2111.02080","source_version":6,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:38:46Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xdcx99tGi1GhB558LXyL887PRaMqFosxVZP2Jl2at3S/wacxt1ELtbl+DCNu3mwxMp20XkRl0sE7VX9jXHYjBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T02:30:57.184719Z"},"content_sha256":"6dec5a03a2a1c160de055ea223c4d14586b22f387517728b3ac48f10f659c83f","schema_version":"1.0","event_id":"sha256:6dec5a03a2a1c160de055ea223c4d14586b22f387517728b3ac48f10f659c83f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:VPSR4ZICWER5DNM3ZWX4JTRFRJ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"An Explanation of In-context Learning as Implicit Bayesian Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Large language models perform in-context learning by implicitly inferring latent concepts that explain coherence in both pretraining data and prompt examples.","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Aditi Raghunathan, Percy Liang, Sang Michael Xie, Tengyu Ma","submitted_at":"2021-11-03T09:12:33Z","abstract_excerpt":"Large language models (LMs) such as GPT-3 have the surprising ability to do in-context learning, where the model learns to do a downstream task simply by conditioning on a prompt consisting of input-output examples. The LM learns from these examples without being explicitly pretrained to learn. Thus, it is unclear what enables in-context learning. In this paper, we study how in-context learning can emerge when pretraining documents have long-range coherence. Here, the LM must infer a latent document-level concept to generate coherent next tokens during pretraining. At test time, in-context lea"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We prove when this occurs despite a distribution mismatch between prompts and pretraining data in a setting where the pretraining distribution is a mixture of HMMs.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That real-world pretraining corpora exhibit long-range coherence driven by latent document-level concepts that can be adequately captured by a mixture-of-HMMs generative process, and that this mechanism dominates the in-context behavior of large-scale models trained on messy web data.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"In-context learning emerges as implicit Bayesian inference of latent concepts when pretraining data has long-range coherence, proven for mixture-of-HMM distributions and replicated on the synthetic GINC dataset.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Large language models perform in-context learning by implicitly inferring latent concepts that explain coherence in both pretraining data and prompt examples.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"ad87445a340020ea25925565e642fc90cd2d7a1b8a3a0a4ad17fe215d665560a"},"source":{"id":"2111.02080","kind":"arxiv","version":6},"verdict":{"id":"4e9aa3e5-95e6-49a4-a39f-c3eda946d3df","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T22:18:23.706449Z","strongest_claim":"We prove when this occurs despite a distribution mismatch between prompts and pretraining data in a setting where the pretraining distribution is a mixture of HMMs.","one_line_summary":"In-context learning emerges as implicit Bayesian inference of latent concepts when pretraining data has long-range coherence, proven for mixture-of-HMM distributions and replicated on the synthetic GINC dataset.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That real-world pretraining corpora exhibit long-range coherence driven by latent document-level concepts that can be adequately captured by a mixture-of-HMMs generative process, and that this mechanism dominates the in-context behavior of large-scale models trained on messy web data.","pith_extraction_headline":"Large language models perform in-context learning by implicitly inferring latent concepts that explain coherence in both pretraining data and prompt examples."},"references":{"count":300,"sample":[{"doi":"","year":1966,"title":"Statistical inference for probabilistic functions of finite state markov chains","work_id":"7224cac5-ed47-4032-b64c-473b3a111b50","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2003,"title":"Blei, Andrew Ng, and M","work_id":"9626107e-2b80-43c8-a7fa-a52236887ed4","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2005,"title":"Language Models are Few-Shot Learners","work_id":"214732c0-2edd-44a0-af9e-28184a2b8279","ref_index":3,"cited_arxiv_id":"2005.14165","is_internal_anchor":true},{"doi":"","year":2020,"title":"Le, and Christopher D","work_id":"eeae921c-36f3-45b9-82ec-48c0ad725757","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1977,"title":"A. P. Dempster, Laird N. M., and Rubin D. B. Maximum likelihood from incomplete data via the EM algorithm. Journal of the Royal Statistical Society: Series B, 39 0 (1): 0 1--38, 1977","work_id":"cd7ad5ee-3f69-45e9-a1c4-9f5fdf6f5d9d","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":300,"snapshot_sha256":"7bb962617aeb35410bc72bce51a63365d8ead76373f80f95fa125fcc4c237a12","internal_anchors":7},"formal_canon":{"evidence_count":2,"snapshot_sha256":"40ce0e32a6fa138e01c6a9a65554250320e68e88450f32a46c2f3a4be0865985"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"4e9aa3e5-95e6-49a4-a39f-c3eda946d3df"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:38:46Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"YhZSMJUZ2huMYqbB31s5U0mWlrM3PoMGBwZwVZIaq7m6MK2xjHZ96kat5O70J7v0yhRLU3m393ByPn+rqxi3AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T02:30:57.185864Z"},"content_sha256":"f4603f789db25309a801892e9e3b86b2dc490c00bb8ce2b2d6e6c6254e8133cf","schema_version":"1.0","event_id":"sha256:f4603f789db25309a801892e9e3b86b2dc490c00bb8ce2b2d6e6c6254e8133cf"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/VPSR4ZICWER5DNM3ZWX4JTRFRJ/bundle.json","state_url":"https://pith.science/pith/VPSR4ZICWER5DNM3ZWX4JTRFRJ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/VPSR4ZICWER5DNM3ZWX4JTRFRJ/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-31T02:30:57Z","links":{"resolver":"https://pith.science/pith/VPSR4ZICWER5DNM3ZWX4JTRFRJ","bundle":"https://pith.science/pith/VPSR4ZICWER5DNM3ZWX4JTRFRJ/bundle.json","state":"https://pith.science/pith/VPSR4ZICWER5DNM3ZWX4JTRFRJ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/VPSR4ZICWER5DNM3ZWX4JTRFRJ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:VPSR4ZICWER5DNM3ZWX4JTRFRJ","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"ece93fb738dd7f046125c81a57fa34c43199dcc6730005f0994bac4c271d5308","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-11-03T09:12:33Z","title_canon_sha256":"d11121fff8b4527ebb3daba0fca9697ace21e7716128ef726203c70bc0afe3a2"},"schema_version":"1.0","source":{"id":"2111.02080","kind":"arxiv","version":6}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2111.02080","created_at":"2026-05-17T23:38:46Z"},{"alias_kind":"arxiv_version","alias_value":"2111.02080v6","created_at":"2026-05-17T23:38:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2111.02080","created_at":"2026-05-17T23:38:46Z"},{"alias_kind":"pith_short_12","alias_value":"VPSR4ZICWER5","created_at":"2026-05-18T12:33:33Z"},{"alias_kind":"pith_short_16","alias_value":"VPSR4ZICWER5DNM3","created_at":"2026-05-18T12:33:33Z"},{"alias_kind":"pith_short_8","alias_value":"VPSR4ZIC","created_at":"2026-05-18T12:33:33Z"}],"graph_snapshots":[{"event_id":"sha256:f4603f789db25309a801892e9e3b86b2dc490c00bb8ce2b2d6e6c6254e8133cf","target":"graph","created_at":"2026-05-17T23:38:46Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"We prove when this occurs despite a distribution mismatch between prompts and pretraining data in a setting where the pretraining distribution is a mixture of HMMs."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That real-world pretraining corpora exhibit long-range coherence driven by latent document-level concepts that can be adequately captured by a mixture-of-HMMs generative process, and that this mechanism dominates the in-context behavior of large-scale models trained on messy web data."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"In-context learning emerges as implicit Bayesian inference of latent concepts when pretraining data has long-range coherence, proven for mixture-of-HMM distributions and replicated on the synthetic GINC dataset."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Large language models perform in-context learning by implicitly inferring latent concepts that explain coherence in both pretraining data and prompt examples."}],"snapshot_sha256":"ad87445a340020ea25925565e642fc90cd2d7a1b8a3a0a4ad17fe215d665560a"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"40ce0e32a6fa138e01c6a9a65554250320e68e88450f32a46c2f3a4be0865985"},"paper":{"abstract_excerpt":"Large language models (LMs) such as GPT-3 have the surprising ability to do in-context learning, where the model learns to do a downstream task simply by conditioning on a prompt consisting of input-output examples. The LM learns from these examples without being explicitly pretrained to learn. Thus, it is unclear what enables in-context learning. In this paper, we study how in-context learning can emerge when pretraining documents have long-range coherence. Here, the LM must infer a latent document-level concept to generate coherent next tokens during pretraining. At test time, in-context lea","authors_text":"Aditi Raghunathan, Percy Liang, Sang Michael Xie, Tengyu Ma","cross_cats":["cs.LG"],"headline":"Large language models perform in-context learning by implicitly inferring latent concepts that explain coherence in both pretraining data and prompt examples.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-11-03T09:12:33Z","title":"An Explanation of In-context Learning as Implicit Bayesian Inference"},"references":{"count":300,"internal_anchors":7,"resolved_work":300,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Statistical inference for probabilistic functions of finite state markov chains","work_id":"7224cac5-ed47-4032-b64c-473b3a111b50","year":1966},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Blei, Andrew Ng, and M","work_id":"9626107e-2b80-43c8-a7fa-a52236887ed4","year":2003},{"cited_arxiv_id":"2005.14165","doi":"","is_internal_anchor":true,"ref_index":3,"title":"Language Models are Few-Shot Learners","work_id":"214732c0-2edd-44a0-af9e-28184a2b8279","year":2005},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Le, and Christopher D","work_id":"eeae921c-36f3-45b9-82ec-48c0ad725757","year":2020},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"A. P. Dempster, Laird N. M., and Rubin D. B. Maximum likelihood from incomplete data via the EM algorithm. Journal of the Royal Statistical Society: Series B, 39 0 (1): 0 1--38, 1977","work_id":"cd7ad5ee-3f69-45e9-a1c4-9f5fdf6f5d9d","year":1977}],"snapshot_sha256":"7bb962617aeb35410bc72bce51a63365d8ead76373f80f95fa125fcc4c237a12"},"source":{"id":"2111.02080","kind":"arxiv","version":6},"verdict":{"created_at":"2026-05-16T22:18:23.706449Z","id":"4e9aa3e5-95e6-49a4-a39f-c3eda946d3df","model_set":{"reader":"grok-4.3"},"one_line_summary":"In-context learning emerges as implicit Bayesian inference of latent concepts when pretraining data has long-range coherence, proven for mixture-of-HMM distributions and replicated on the synthetic GINC dataset.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Large language models perform in-context learning by implicitly inferring latent concepts that explain coherence in both pretraining data and prompt examples.","strongest_claim":"We prove when this occurs despite a distribution mismatch between prompts and pretraining data in a setting where the pretraining distribution is a mixture of HMMs.","weakest_assumption":"That real-world pretraining corpora exhibit long-range coherence driven by latent document-level concepts that can be adequately captured by a mixture-of-HMMs generative process, and that this mechanism dominates the in-context behavior of large-scale models trained on messy web data."}},"verdict_id":"4e9aa3e5-95e6-49a4-a39f-c3eda946d3df"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:6dec5a03a2a1c160de055ea223c4d14586b22f387517728b3ac48f10f659c83f","target":"record","created_at":"2026-05-17T23:38:46Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"ece93fb738dd7f046125c81a57fa34c43199dcc6730005f0994bac4c271d5308","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-11-03T09:12:33Z","title_canon_sha256":"d11121fff8b4527ebb3daba0fca9697ace21e7716128ef726203c70bc0afe3a2"},"schema_version":"1.0","source":{"id":"2111.02080","kind":"arxiv","version":6}},"canonical_sha256":"abe51e6502b123d1b59bcdafc4ce258a748628095a8fba3c262f049ff320956a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"abe51e6502b123d1b59bcdafc4ce258a748628095a8fba3c262f049ff320956a","first_computed_at":"2026-05-17T23:38:46.404101Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:38:46.404101Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"OrzOJeld4FxePDZPLH4TxbHTvtLJLKQLSIoJs8dF+mk+p42dqTFl0PLpky+KLEdZxnTt+4kSc7y+QDGlGKiZCw==","signature_status":"signed_v1","signed_at":"2026-05-17T23:38:46.404663Z","signed_message":"canonical_sha256_bytes"},"source_id":"2111.02080","source_kind":"arxiv","source_version":6}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6dec5a03a2a1c160de055ea223c4d14586b22f387517728b3ac48f10f659c83f","sha256:f4603f789db25309a801892e9e3b86b2dc490c00bb8ce2b2d6e6c6254e8133cf"],"state_sha256":"98e403be698d725b3fe29ddfe37ac2cf3feade9c55571abed15834c3db985a35"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FsABDuv3v574hPiY5QRcPdVSy24z7aaGW5gqV3VJnqJbROtBElwlkxlkIilCanMp3O/SlwdwXPxBKPICLQrSBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-31T02:30:57.190561Z","bundle_sha256":"1101b4fc8514f65a7c6ff1f176c45ceb5759ea749f7254155a00c662f3b65c12"}}