{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:T7GIUBKQUK5RUDOTKZARBNKJ7G","short_pith_number":"pith:T7GIUBKQ","schema_version":"1.0","canonical_sha256":"9fcc8a0550a2bb1a0dd3564110b549f9bc923ec61ac346c416b8d0af0e9ff166","source":{"kind":"arxiv","id":"2606.20544","version":1},"attestation_state":"computed","paper":{"title":"Toward Calibrated Mixture-of-Experts Under Distribution Shift","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Anqi Liu, Drew Prinster, Gina Wong, Rama Chellappa, Suchi Saria","submitted_at":"2026-06-18T17:55:00Z","abstract_excerpt":"Calibration aligns a model's predictive uncertainty with the frequencies of its empirical outcomes and is important for understanding and trusting reported probabilities. Recent work shows that enforcing calibration at the level of individual predictors can improve ensemble accuracy and calibration, with mixture-of-experts (MoE) models showing strong empirical improvements in particular; however, the conditions under which calibration helps MoE are not well understood. In this work, we study how MoE models behave under distribution shift, focusing on how routing mechanisms interact with expert"},"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":"2606.20544","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-18T17:55:00Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"14dff42aa2600096442693bce53cf16ef79f91d7c80020773b5e5711c509515f","abstract_canon_sha256":"a35d373289ab6c34d613e6e706d93f99cd974244d1d6732ce65519f489c86ebd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:13:14.919659Z","signature_b64":"gHYS6KsmwlTaTiaBGQT0kDFOUT1GFMOZKbNiCXomXbQYBvI6/bS3oiDCKh3m1/I6LDlMNHYfTv8jOnLvedbQCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9fcc8a0550a2bb1a0dd3564110b549f9bc923ec61ac346c416b8d0af0e9ff166","last_reissued_at":"2026-06-19T16:13:14.919300Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:13:14.919300Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Toward Calibrated Mixture-of-Experts Under Distribution Shift","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Anqi Liu, Drew Prinster, Gina Wong, Rama Chellappa, Suchi Saria","submitted_at":"2026-06-18T17:55:00Z","abstract_excerpt":"Calibration aligns a model's predictive uncertainty with the frequencies of its empirical outcomes and is important for understanding and trusting reported probabilities. Recent work shows that enforcing calibration at the level of individual predictors can improve ensemble accuracy and calibration, with mixture-of-experts (MoE) models showing strong empirical improvements in particular; however, the conditions under which calibration helps MoE are not well understood. In this work, we study how MoE models behave under distribution shift, focusing on how routing mechanisms interact with expert"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.20544","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/2606.20544/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":"2606.20544","created_at":"2026-06-19T16:13:14.919373+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.20544v1","created_at":"2026-06-19T16:13:14.919373+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.20544","created_at":"2026-06-19T16:13:14.919373+00:00"},{"alias_kind":"pith_short_12","alias_value":"T7GIUBKQUK5R","created_at":"2026-06-19T16:13:14.919373+00:00"},{"alias_kind":"pith_short_16","alias_value":"T7GIUBKQUK5RUDOT","created_at":"2026-06-19T16:13:14.919373+00:00"},{"alias_kind":"pith_short_8","alias_value":"T7GIUBKQ","created_at":"2026-06-19T16:13:14.919373+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/T7GIUBKQUK5RUDOTKZARBNKJ7G","json":"https://pith.science/pith/T7GIUBKQUK5RUDOTKZARBNKJ7G.json","graph_json":"https://pith.science/api/pith-number/T7GIUBKQUK5RUDOTKZARBNKJ7G/graph.json","events_json":"https://pith.science/api/pith-number/T7GIUBKQUK5RUDOTKZARBNKJ7G/events.json","paper":"https://pith.science/paper/T7GIUBKQ"},"agent_actions":{"view_html":"https://pith.science/pith/T7GIUBKQUK5RUDOTKZARBNKJ7G","download_json":"https://pith.science/pith/T7GIUBKQUK5RUDOTKZARBNKJ7G.json","view_paper":"https://pith.science/paper/T7GIUBKQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.20544&json=true","fetch_graph":"https://pith.science/api/pith-number/T7GIUBKQUK5RUDOTKZARBNKJ7G/graph.json","fetch_events":"https://pith.science/api/pith-number/T7GIUBKQUK5RUDOTKZARBNKJ7G/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/T7GIUBKQUK5RUDOTKZARBNKJ7G/action/timestamp_anchor","attest_storage":"https://pith.science/pith/T7GIUBKQUK5RUDOTKZARBNKJ7G/action/storage_attestation","attest_author":"https://pith.science/pith/T7GIUBKQUK5RUDOTKZARBNKJ7G/action/author_attestation","sign_citation":"https://pith.science/pith/T7GIUBKQUK5RUDOTKZARBNKJ7G/action/citation_signature","submit_replication":"https://pith.science/pith/T7GIUBKQUK5RUDOTKZARBNKJ7G/action/replication_record"}},"created_at":"2026-06-19T16:13:14.919373+00:00","updated_at":"2026-06-19T16:13:14.919373+00:00"}