{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:VEZEJEUNV7CWVSQMIOW3JB7GDO","short_pith_number":"pith:VEZEJEUN","schema_version":"1.0","canonical_sha256":"a93244928dafc56aca0c43adb487e61b997c925828d7bc8f5ddcb9434e485735","source":{"kind":"arxiv","id":"2606.21749","version":1},"attestation_state":"computed","paper":{"title":"Quantile Adaptive Temperature Scaling for Confidence Calibration","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Ismail Ben Ayed, Jose Dolz, Leo Fillioux, Omprakash Chakraborty","submitted_at":"2026-06-19T21:06:11Z","abstract_excerpt":"Deep neural networks often produce poorly calibrated confidence estimates, overstating their certainty even when predictions are incorrect. Temperature Scaling remains the most widely used posthoc calibration method due to its simplicity and effectiveness, yet its global, uniform rescaling of logits fails to correct the highly heterogeneous structure of miscalibration observed across the confidence spectrum. In particular, the largest correctness confidence discrepancies arise in different quantile regions depending on the setting, low confidence predictions, where uncertainty matters most, te"},"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":"2606.21749","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-19T21:06:11Z","cross_cats_sorted":[],"title_canon_sha256":"e8a7b28968e791726aedcf9c9320893b44cea5a1101637d4d4e5aa3d6c292381","abstract_canon_sha256":"e1c5ad123fb76d0955f2e71942786fbf62f908fcd0b6740e87da506e682d21c1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T01:13:21.439217Z","signature_b64":"lSxPqr1nTWQba3wkbKeoyDgc+uNpZdw7wAH5FYbBTECkgjVVoAf9GUB1bh4yNtCZ+ZvYbr4vUpjp3GEy2fheBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a93244928dafc56aca0c43adb487e61b997c925828d7bc8f5ddcb9434e485735","last_reissued_at":"2026-06-23T01:13:21.438769Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T01:13:21.438769Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Quantile Adaptive Temperature Scaling for Confidence Calibration","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Ismail Ben Ayed, Jose Dolz, Leo Fillioux, Omprakash Chakraborty","submitted_at":"2026-06-19T21:06:11Z","abstract_excerpt":"Deep neural networks often produce poorly calibrated confidence estimates, overstating their certainty even when predictions are incorrect. Temperature Scaling remains the most widely used posthoc calibration method due to its simplicity and effectiveness, yet its global, uniform rescaling of logits fails to correct the highly heterogeneous structure of miscalibration observed across the confidence spectrum. In particular, the largest correctness confidence discrepancies arise in different quantile regions depending on the setting, low confidence predictions, where uncertainty matters most, te"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.21749","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/2606.21749/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":"2606.21749","created_at":"2026-06-23T01:13:21.438836+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.21749v1","created_at":"2026-06-23T01:13:21.438836+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.21749","created_at":"2026-06-23T01:13:21.438836+00:00"},{"alias_kind":"pith_short_12","alias_value":"VEZEJEUNV7CW","created_at":"2026-06-23T01:13:21.438836+00:00"},{"alias_kind":"pith_short_16","alias_value":"VEZEJEUNV7CWVSQM","created_at":"2026-06-23T01:13:21.438836+00:00"},{"alias_kind":"pith_short_8","alias_value":"VEZEJEUN","created_at":"2026-06-23T01:13:21.438836+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/VEZEJEUNV7CWVSQMIOW3JB7GDO","json":"https://pith.science/pith/VEZEJEUNV7CWVSQMIOW3JB7GDO.json","graph_json":"https://pith.science/api/pith-number/VEZEJEUNV7CWVSQMIOW3JB7GDO/graph.json","events_json":"https://pith.science/api/pith-number/VEZEJEUNV7CWVSQMIOW3JB7GDO/events.json","paper":"https://pith.science/paper/VEZEJEUN"},"agent_actions":{"view_html":"https://pith.science/pith/VEZEJEUNV7CWVSQMIOW3JB7GDO","download_json":"https://pith.science/pith/VEZEJEUNV7CWVSQMIOW3JB7GDO.json","view_paper":"https://pith.science/paper/VEZEJEUN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.21749&json=true","fetch_graph":"https://pith.science/api/pith-number/VEZEJEUNV7CWVSQMIOW3JB7GDO/graph.json","fetch_events":"https://pith.science/api/pith-number/VEZEJEUNV7CWVSQMIOW3JB7GDO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VEZEJEUNV7CWVSQMIOW3JB7GDO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VEZEJEUNV7CWVSQMIOW3JB7GDO/action/storage_attestation","attest_author":"https://pith.science/pith/VEZEJEUNV7CWVSQMIOW3JB7GDO/action/author_attestation","sign_citation":"https://pith.science/pith/VEZEJEUNV7CWVSQMIOW3JB7GDO/action/citation_signature","submit_replication":"https://pith.science/pith/VEZEJEUNV7CWVSQMIOW3JB7GDO/action/replication_record"}},"created_at":"2026-06-23T01:13:21.438836+00:00","updated_at":"2026-06-23T01:13:21.438836+00:00"}