{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:SBX7HONAPFLJGUPZSLIRCFKLV6","short_pith_number":"pith:SBX7HONA","schema_version":"1.0","canonical_sha256":"906ff3b9a079569351f992d111154baf96d5314a8071f754ac4815bfbcb0eb26","source":{"kind":"arxiv","id":"2603.09453","version":3},"attestation_state":"computed","paper":{"title":"Variational Routing: A Scalable Bayesian Framework for Calibrated Mixture-of-Experts Transformers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Albus Yizhuo Li, Matthew Wicker","submitted_at":"2026-03-10T10:07:53Z","abstract_excerpt":"Foundation models are increasingly being deployed in contexts where understanding the uncertainty of their outputs is critical to ensuring responsible deployment. While Bayesian methods offer a principled approach to uncertainty quantification, their computational overhead renders their use impractical for training or inference at foundation model scale. State-of-the-art models achieve parameter counts in the trillions through carefully engineered sparsity including Mixture-of-Experts (MoE) layers. In this work, we demonstrate calibrated uncertainty at scale by introducing Variational Mixture-"},"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":"2603.09453","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-03-10T10:07:53Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"eb9dd7882dc3dfe2d75c6cd08fd51d0c636e13f1d05313fc91cf3efaf11c50b7","abstract_canon_sha256":"d71e67209f77b17399b537786c0a203c8e8ef93164ca63c3b974420f35f3e685"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-01T01:02:36.657361Z","signature_b64":"D7290LQilf1l3Bb009tyvsFSSPuWhH1glT1x5QxeSpjJ8U+jzFDWZuUwdz0i+2jO6g0l19Sq9ILFISgTMw69DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"906ff3b9a079569351f992d111154baf96d5314a8071f754ac4815bfbcb0eb26","last_reissued_at":"2026-06-01T01:02:36.656163Z","signature_status":"signed_v1","first_computed_at":"2026-06-01T01:02:36.656163Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Variational Routing: A Scalable Bayesian Framework for Calibrated Mixture-of-Experts Transformers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Albus Yizhuo Li, Matthew Wicker","submitted_at":"2026-03-10T10:07:53Z","abstract_excerpt":"Foundation models are increasingly being deployed in contexts where understanding the uncertainty of their outputs is critical to ensuring responsible deployment. While Bayesian methods offer a principled approach to uncertainty quantification, their computational overhead renders their use impractical for training or inference at foundation model scale. State-of-the-art models achieve parameter counts in the trillions through carefully engineered sparsity including Mixture-of-Experts (MoE) layers. In this work, we demonstrate calibrated uncertainty at scale by introducing Variational Mixture-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.09453","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/2603.09453/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":"2603.09453","created_at":"2026-06-01T01:02:36.656352+00:00"},{"alias_kind":"arxiv_version","alias_value":"2603.09453v3","created_at":"2026-06-01T01:02:36.656352+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.09453","created_at":"2026-06-01T01:02:36.656352+00:00"},{"alias_kind":"pith_short_12","alias_value":"SBX7HONAPFLJ","created_at":"2026-06-01T01:02:36.656352+00:00"},{"alias_kind":"pith_short_16","alias_value":"SBX7HONAPFLJGUPZ","created_at":"2026-06-01T01:02:36.656352+00:00"},{"alias_kind":"pith_short_8","alias_value":"SBX7HONA","created_at":"2026-06-01T01:02:36.656352+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.00604","citing_title":"Affinity Is Not Enough: Recovering the Free Energy Principle in Mixture-of-Experts","ref_index":33,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SBX7HONAPFLJGUPZSLIRCFKLV6","json":"https://pith.science/pith/SBX7HONAPFLJGUPZSLIRCFKLV6.json","graph_json":"https://pith.science/api/pith-number/SBX7HONAPFLJGUPZSLIRCFKLV6/graph.json","events_json":"https://pith.science/api/pith-number/SBX7HONAPFLJGUPZSLIRCFKLV6/events.json","paper":"https://pith.science/paper/SBX7HONA"},"agent_actions":{"view_html":"https://pith.science/pith/SBX7HONAPFLJGUPZSLIRCFKLV6","download_json":"https://pith.science/pith/SBX7HONAPFLJGUPZSLIRCFKLV6.json","view_paper":"https://pith.science/paper/SBX7HONA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2603.09453&json=true","fetch_graph":"https://pith.science/api/pith-number/SBX7HONAPFLJGUPZSLIRCFKLV6/graph.json","fetch_events":"https://pith.science/api/pith-number/SBX7HONAPFLJGUPZSLIRCFKLV6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SBX7HONAPFLJGUPZSLIRCFKLV6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SBX7HONAPFLJGUPZSLIRCFKLV6/action/storage_attestation","attest_author":"https://pith.science/pith/SBX7HONAPFLJGUPZSLIRCFKLV6/action/author_attestation","sign_citation":"https://pith.science/pith/SBX7HONAPFLJGUPZSLIRCFKLV6/action/citation_signature","submit_replication":"https://pith.science/pith/SBX7HONAPFLJGUPZSLIRCFKLV6/action/replication_record"}},"created_at":"2026-06-01T01:02:36.656352+00:00","updated_at":"2026-06-01T01:02:36.656352+00:00"}