{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:PHCRSYEASEIX2M3PFJUIGIQKTI","short_pith_number":"pith:PHCRSYEA","schema_version":"1.0","canonical_sha256":"79c519608091117d336f2a6883220a9a2a4c0d2b59f215aa7976a571d9de024a","source":{"kind":"arxiv","id":"2206.00826","version":1},"attestation_state":"computed","paper":{"title":"BayesFormer: Transformer with Uncertainty Estimation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Han Fang, Karthik Abinav Sankararaman, Sinong Wang","submitted_at":"2022-06-02T01:54:58Z","abstract_excerpt":"Transformer has become ubiquitous due to its dominant performance in various NLP and image processing tasks. However, it lacks understanding of how to generate mathematically grounded uncertainty estimates for transformer architectures. Models equipped with such uncertainty estimates can typically improve predictive performance, make networks robust, avoid over-fitting and used as acquisition function in active learning. In this paper, we introduce BayesFormer, a Transformer model with dropouts designed by Bayesian theory. We proposed a new theoretical framework to extend the approximate varia"},"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":"2206.00826","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-06-02T01:54:58Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"0959977879353eb925dbd2b843fb0c47cff4c0981d0d91ffb35bc2132ce6a4b6","abstract_canon_sha256":"f3b69d5bef2575c15cdabb7d191f150885b205d347bdd94b03f06c0be9af4ec3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:28:34.096182Z","signature_b64":"QJokdAec0UsHjOizZ+Klpq2gEK/a/jp7Ed6M97PK6Rn/keDakObn6wCRbkuIFclUi6BcBgyWu4rtyrbNmppADA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"79c519608091117d336f2a6883220a9a2a4c0d2b59f215aa7976a571d9de024a","last_reissued_at":"2026-07-05T04:28:34.095790Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:28:34.095790Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"BayesFormer: Transformer with Uncertainty Estimation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Han Fang, Karthik Abinav Sankararaman, Sinong Wang","submitted_at":"2022-06-02T01:54:58Z","abstract_excerpt":"Transformer has become ubiquitous due to its dominant performance in various NLP and image processing tasks. However, it lacks understanding of how to generate mathematically grounded uncertainty estimates for transformer architectures. Models equipped with such uncertainty estimates can typically improve predictive performance, make networks robust, avoid over-fitting and used as acquisition function in active learning. In this paper, we introduce BayesFormer, a Transformer model with dropouts designed by Bayesian theory. We proposed a new theoretical framework to extend the approximate varia"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2206.00826","kind":"arxiv","version":1},"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/2206.00826/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":"2206.00826","created_at":"2026-07-05T04:28:34.095845+00:00"},{"alias_kind":"arxiv_version","alias_value":"2206.00826v1","created_at":"2026-07-05T04:28:34.095845+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.00826","created_at":"2026-07-05T04:28:34.095845+00:00"},{"alias_kind":"pith_short_12","alias_value":"PHCRSYEASEIX","created_at":"2026-07-05T04:28:34.095845+00:00"},{"alias_kind":"pith_short_16","alias_value":"PHCRSYEASEIX2M3P","created_at":"2026-07-05T04:28:34.095845+00:00"},{"alias_kind":"pith_short_8","alias_value":"PHCRSYEA","created_at":"2026-07-05T04:28:34.095845+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2506.22809","citing_title":"Learning Adapter Rank via Symmetry Breaking","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11007","citing_title":"The Transformer as a Polar State Estimator","ref_index":90,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03614","citing_title":"Uncertainty Estimation in Instance Segmentation of Affordances via Bayesian Visual Transformers","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12513","citing_title":"Agentic Control in Variational Language Models","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PHCRSYEASEIX2M3PFJUIGIQKTI","json":"https://pith.science/pith/PHCRSYEASEIX2M3PFJUIGIQKTI.json","graph_json":"https://pith.science/api/pith-number/PHCRSYEASEIX2M3PFJUIGIQKTI/graph.json","events_json":"https://pith.science/api/pith-number/PHCRSYEASEIX2M3PFJUIGIQKTI/events.json","paper":"https://pith.science/paper/PHCRSYEA"},"agent_actions":{"view_html":"https://pith.science/pith/PHCRSYEASEIX2M3PFJUIGIQKTI","download_json":"https://pith.science/pith/PHCRSYEASEIX2M3PFJUIGIQKTI.json","view_paper":"https://pith.science/paper/PHCRSYEA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2206.00826&json=true","fetch_graph":"https://pith.science/api/pith-number/PHCRSYEASEIX2M3PFJUIGIQKTI/graph.json","fetch_events":"https://pith.science/api/pith-number/PHCRSYEASEIX2M3PFJUIGIQKTI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PHCRSYEASEIX2M3PFJUIGIQKTI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PHCRSYEASEIX2M3PFJUIGIQKTI/action/storage_attestation","attest_author":"https://pith.science/pith/PHCRSYEASEIX2M3PFJUIGIQKTI/action/author_attestation","sign_citation":"https://pith.science/pith/PHCRSYEASEIX2M3PFJUIGIQKTI/action/citation_signature","submit_replication":"https://pith.science/pith/PHCRSYEASEIX2M3PFJUIGIQKTI/action/replication_record"}},"created_at":"2026-07-05T04:28:34.095845+00:00","updated_at":"2026-07-05T04:28:34.095845+00:00"}