{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:PB4X3W72MZVPIKW4PK3CTSHTU7","short_pith_number":"pith:PB4X3W72","canonical_record":{"source":{"id":"2605.30188","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-28T16:31:36Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"68292f4ae1f2df4020a4b6cf901e144450d6bdbe954f63a4218fee9c6f18fde5","abstract_canon_sha256":"297ccb391ef23342c3cc7b8b3c5cebcdd1f6206f52af4f38e5fa940ff2f6a49e"},"schema_version":"1.0"},"canonical_sha256":"78797ddbfa666af42adc7ab629c8f3a7d3a0255d76e6d910c15d13981ac4749f","source":{"kind":"arxiv","id":"2605.30188","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.30188","created_at":"2026-05-29T02:06:12Z"},{"alias_kind":"arxiv_version","alias_value":"2605.30188v1","created_at":"2026-05-29T02:06:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.30188","created_at":"2026-05-29T02:06:12Z"},{"alias_kind":"pith_short_12","alias_value":"PB4X3W72MZVP","created_at":"2026-05-29T02:06:12Z"},{"alias_kind":"pith_short_16","alias_value":"PB4X3W72MZVPIKW4","created_at":"2026-05-29T02:06:12Z"},{"alias_kind":"pith_short_8","alias_value":"PB4X3W72","created_at":"2026-05-29T02:06:12Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:PB4X3W72MZVPIKW4PK3CTSHTU7","target":"record","payload":{"canonical_record":{"source":{"id":"2605.30188","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-28T16:31:36Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"68292f4ae1f2df4020a4b6cf901e144450d6bdbe954f63a4218fee9c6f18fde5","abstract_canon_sha256":"297ccb391ef23342c3cc7b8b3c5cebcdd1f6206f52af4f38e5fa940ff2f6a49e"},"schema_version":"1.0"},"canonical_sha256":"78797ddbfa666af42adc7ab629c8f3a7d3a0255d76e6d910c15d13981ac4749f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T02:06:12.348220Z","signature_b64":"d4VyZCsCOA6NZgAvrgbY2MQw9Mx3y+6/hj/hwM/h6LY+D6iYmESKSBRPmMXUEU9UHzZ4f7I74xqI07yXd5B7Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"78797ddbfa666af42adc7ab629c8f3a7d3a0255d76e6d910c15d13981ac4749f","last_reissued_at":"2026-05-29T02:06:12.347812Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T02:06:12.347812Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.30188","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-29T02:06:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tyi7C0nmUvgL+hdqX88kgEwqt16e78gzJQAHEMjXACFGE8wLGRl/a5eQLXBivU9pTMqqp8eqmKxibQl/aiZYBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T00:25:20.008783Z"},"content_sha256":"f6e3148347d781fd7f70a37616582cde2567c260f4ac16f03df9ae9a1fa56032","schema_version":"1.0","event_id":"sha256:f6e3148347d781fd7f70a37616582cde2567c260f4ac16f03df9ae9a1fa56032"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:PB4X3W72MZVPIKW4PK3CTSHTU7","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"CalArena: A Large-Scale Post-Hoc Calibration Benchmark","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"David Holzm\\\"uller, Eug\\`ene Berta, Francis Bach, Michael I. Jordan","submitted_at":"2026-05-28T16:31:36Z","abstract_excerpt":"Reliable probability estimates are critical in many machine learning applications, yet modern classifiers are often poorly calibrated. Post-hoc calibration provides a simple and widely used solution, but the large number of proposed methods, combined with small-scale and inconsistent evaluations, makes it difficult to determine which approaches are truly effective in practice. We introduce a large-scale, standardized benchmark for post-hoc calibration, covering nearly 2000 experiments across tabular and computer vision tasks, including binary, multiclass, and large-scale classification setting"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.30188","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/2605.30188/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-29T02:06:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1/Ihdn7deIorLOnZZLiDhgcNDLgO3UEbBt2U7ihxgH1bHuqC079V+8tDc6lqINB9+mBqwklk1ipMKO69roKIBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T00:25:20.009156Z"},"content_sha256":"d96f6a8a8d6593636738e1fabd6e6914313304c75cf2c7b37574b59f534d7c94","schema_version":"1.0","event_id":"sha256:d96f6a8a8d6593636738e1fabd6e6914313304c75cf2c7b37574b59f534d7c94"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/PB4X3W72MZVPIKW4PK3CTSHTU7/bundle.json","state_url":"https://pith.science/pith/PB4X3W72MZVPIKW4PK3CTSHTU7/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/PB4X3W72MZVPIKW4PK3CTSHTU7/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":"2026-05-31T00:25:20Z","links":{"resolver":"https://pith.science/pith/PB4X3W72MZVPIKW4PK3CTSHTU7","bundle":"https://pith.science/pith/PB4X3W72MZVPIKW4PK3CTSHTU7/bundle.json","state":"https://pith.science/pith/PB4X3W72MZVPIKW4PK3CTSHTU7/state.json","well_known_bundle":"https://pith.science/.well-known/pith/PB4X3W72MZVPIKW4PK3CTSHTU7/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:PB4X3W72MZVPIKW4PK3CTSHTU7","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":"297ccb391ef23342c3cc7b8b3c5cebcdd1f6206f52af4f38e5fa940ff2f6a49e","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-28T16:31:36Z","title_canon_sha256":"68292f4ae1f2df4020a4b6cf901e144450d6bdbe954f63a4218fee9c6f18fde5"},"schema_version":"1.0","source":{"id":"2605.30188","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.30188","created_at":"2026-05-29T02:06:12Z"},{"alias_kind":"arxiv_version","alias_value":"2605.30188v1","created_at":"2026-05-29T02:06:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.30188","created_at":"2026-05-29T02:06:12Z"},{"alias_kind":"pith_short_12","alias_value":"PB4X3W72MZVP","created_at":"2026-05-29T02:06:12Z"},{"alias_kind":"pith_short_16","alias_value":"PB4X3W72MZVPIKW4","created_at":"2026-05-29T02:06:12Z"},{"alias_kind":"pith_short_8","alias_value":"PB4X3W72","created_at":"2026-05-29T02:06:12Z"}],"graph_snapshots":[{"event_id":"sha256:d96f6a8a8d6593636738e1fabd6e6914313304c75cf2c7b37574b59f534d7c94","target":"graph","created_at":"2026-05-29T02:06:12Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2605.30188/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reliable probability estimates are critical in many machine learning applications, yet modern classifiers are often poorly calibrated. Post-hoc calibration provides a simple and widely used solution, but the large number of proposed methods, combined with small-scale and inconsistent evaluations, makes it difficult to determine which approaches are truly effective in practice. We introduce a large-scale, standardized benchmark for post-hoc calibration, covering nearly 2000 experiments across tabular and computer vision tasks, including binary, multiclass, and large-scale classification setting","authors_text":"David Holzm\\\"uller, Eug\\`ene Berta, Francis Bach, Michael I. Jordan","cross_cats":["cs.AI","stat.ML"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-28T16:31:36Z","title":"CalArena: A Large-Scale Post-Hoc Calibration Benchmark"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.30188","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:f6e3148347d781fd7f70a37616582cde2567c260f4ac16f03df9ae9a1fa56032","target":"record","created_at":"2026-05-29T02:06:12Z","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":"297ccb391ef23342c3cc7b8b3c5cebcdd1f6206f52af4f38e5fa940ff2f6a49e","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-28T16:31:36Z","title_canon_sha256":"68292f4ae1f2df4020a4b6cf901e144450d6bdbe954f63a4218fee9c6f18fde5"},"schema_version":"1.0","source":{"id":"2605.30188","kind":"arxiv","version":1}},"canonical_sha256":"78797ddbfa666af42adc7ab629c8f3a7d3a0255d76e6d910c15d13981ac4749f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"78797ddbfa666af42adc7ab629c8f3a7d3a0255d76e6d910c15d13981ac4749f","first_computed_at":"2026-05-29T02:06:12.347812Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-29T02:06:12.347812Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"d4VyZCsCOA6NZgAvrgbY2MQw9Mx3y+6/hj/hwM/h6LY+D6iYmESKSBRPmMXUEU9UHzZ4f7I74xqI07yXd5B7Dg==","signature_status":"signed_v1","signed_at":"2026-05-29T02:06:12.348220Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.30188","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f6e3148347d781fd7f70a37616582cde2567c260f4ac16f03df9ae9a1fa56032","sha256:d96f6a8a8d6593636738e1fabd6e6914313304c75cf2c7b37574b59f534d7c94"],"state_sha256":"e2f7c5cc1873a339ae6983e9bb23d4a6d75ef2ba5048315b8b23f4de7cdfa33b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"bA89cyiuR3fyQS2kn6xLtRWKn8ShYjK3XezmGPCjxQpp8B93nUdUsoJ6RFA8FLCkzlResD8x4HF5G3XctSz/Ag==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-31T00:25:20.011303Z","bundle_sha256":"4000f03676947732e69d8ef9f68acbc71e522a0abac40bc23057bc0d8fd45428"}}