{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:JAZYA5E3FZNCXK6ROVYQZMAALV","short_pith_number":"pith:JAZYA5E3","canonical_record":{"source":{"id":"2404.07815","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-04-11T14:58:19Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"66abc60555658b80f0feeca70a874c979f1d70d7361dcce6e486f025e728314d","abstract_canon_sha256":"8b678b19412d031a529332ccbadfdeb34f4b8d13f3364b6fe97cb29767b55f85"},"schema_version":"1.0"},"canonical_sha256":"483380749b2e5a2babd175710cb0005d70907822e799bc640d8bd90add861da1","source":{"kind":"arxiv","id":"2404.07815","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.07815","created_at":"2026-07-05T09:15:31Z"},{"alias_kind":"arxiv_version","alias_value":"2404.07815v2","created_at":"2026-07-05T09:15:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.07815","created_at":"2026-07-05T09:15:31Z"},{"alias_kind":"pith_short_12","alias_value":"JAZYA5E3FZNC","created_at":"2026-07-05T09:15:31Z"},{"alias_kind":"pith_short_16","alias_value":"JAZYA5E3FZNCXK6R","created_at":"2026-07-05T09:15:31Z"},{"alias_kind":"pith_short_8","alias_value":"JAZYA5E3","created_at":"2026-07-05T09:15:31Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:JAZYA5E3FZNCXK6ROVYQZMAALV","target":"record","payload":{"canonical_record":{"source":{"id":"2404.07815","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-04-11T14:58:19Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"66abc60555658b80f0feeca70a874c979f1d70d7361dcce6e486f025e728314d","abstract_canon_sha256":"8b678b19412d031a529332ccbadfdeb34f4b8d13f3364b6fe97cb29767b55f85"},"schema_version":"1.0"},"canonical_sha256":"483380749b2e5a2babd175710cb0005d70907822e799bc640d8bd90add861da1","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:15:31.298140Z","signature_b64":"oyH3F3E1mxcXdv1EJR1jRvqgmftSIOMb8wj2/Kk9Gkf/zBtOvPTbWmzO4fW/V5GMRpUxxgvsT2ADn/pcJQD3AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"483380749b2e5a2babd175710cb0005d70907822e799bc640d8bd90add861da1","last_reissued_at":"2026-07-05T09:15:31.297654Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:15:31.297654Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2404.07815","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:15:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UKT34b2kxLx5LEkgQ1PakwVislsMtPo1gDlOMvEaLObMWNHjUzsYUTVGgPriJZa19y7PKOBVshC0IhEeowkYDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T08:53:18.234452Z"},"content_sha256":"872b93bbb867fcada36e64334ed9b9389a967e05c6a85985b5e6acfe59e38d6f","schema_version":"1.0","event_id":"sha256:872b93bbb867fcada36e64334ed9b9389a967e05c6a85985b5e6acfe59e38d6f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:JAZYA5E3FZNCXK6ROVYQZMAALV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Post-Hoc Reversal: Are We Selecting Models Prematurely?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Carlos Guestrin, Mrigank Raman, Rishabh Ranjan, Saurabh Garg, Zachary Lipton","submitted_at":"2024-04-11T14:58:19Z","abstract_excerpt":"Trained models are often composed with post-hoc transforms such as temperature scaling (TS), ensembling and stochastic weight averaging (SWA) to improve performance, robustness, uncertainty estimation, etc. However, such transforms are typically applied only after the base models have already been finalized by standard means. In this paper, we challenge this practice with an extensive empirical study. In particular, we demonstrate a phenomenon that we call post-hoc reversal, where performance trends are reversed after applying post-hoc transforms. This phenomenon is especially prominent in hig"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.07815","kind":"arxiv","version":2},"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/2404.07815/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":"2026-07-05T09:15:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mP1nDjS+TalWXrihDWpDZARVH9Y1Vz1mOSXiSyKv0ttMkRjcylvTP48hyuZs4kJ2i4m5PrK8eG4tLarJVmcwCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T08:53:18.234956Z"},"content_sha256":"daf1d306910da6a95126538eb989591a313e555982f0db33430024243a6e213e","schema_version":"1.0","event_id":"sha256:daf1d306910da6a95126538eb989591a313e555982f0db33430024243a6e213e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/JAZYA5E3FZNCXK6ROVYQZMAALV/bundle.json","state_url":"https://pith.science/pith/JAZYA5E3FZNCXK6ROVYQZMAALV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/JAZYA5E3FZNCXK6ROVYQZMAALV/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":"2026-08-04T08:53:18Z","links":{"resolver":"https://pith.science/pith/JAZYA5E3FZNCXK6ROVYQZMAALV","bundle":"https://pith.science/pith/JAZYA5E3FZNCXK6ROVYQZMAALV/bundle.json","state":"https://pith.science/pith/JAZYA5E3FZNCXK6ROVYQZMAALV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/JAZYA5E3FZNCXK6ROVYQZMAALV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:JAZYA5E3FZNCXK6ROVYQZMAALV","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":"8b678b19412d031a529332ccbadfdeb34f4b8d13f3364b6fe97cb29767b55f85","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-04-11T14:58:19Z","title_canon_sha256":"66abc60555658b80f0feeca70a874c979f1d70d7361dcce6e486f025e728314d"},"schema_version":"1.0","source":{"id":"2404.07815","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.07815","created_at":"2026-07-05T09:15:31Z"},{"alias_kind":"arxiv_version","alias_value":"2404.07815v2","created_at":"2026-07-05T09:15:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.07815","created_at":"2026-07-05T09:15:31Z"},{"alias_kind":"pith_short_12","alias_value":"JAZYA5E3FZNC","created_at":"2026-07-05T09:15:31Z"},{"alias_kind":"pith_short_16","alias_value":"JAZYA5E3FZNCXK6R","created_at":"2026-07-05T09:15:31Z"},{"alias_kind":"pith_short_8","alias_value":"JAZYA5E3","created_at":"2026-07-05T09:15:31Z"}],"graph_snapshots":[{"event_id":"sha256:daf1d306910da6a95126538eb989591a313e555982f0db33430024243a6e213e","target":"graph","created_at":"2026-07-05T09:15:31Z","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/2404.07815/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Trained models are often composed with post-hoc transforms such as temperature scaling (TS), ensembling and stochastic weight averaging (SWA) to improve performance, robustness, uncertainty estimation, etc. However, such transforms are typically applied only after the base models have already been finalized by standard means. In this paper, we challenge this practice with an extensive empirical study. In particular, we demonstrate a phenomenon that we call post-hoc reversal, where performance trends are reversed after applying post-hoc transforms. This phenomenon is especially prominent in hig","authors_text":"Carlos Guestrin, Mrigank Raman, Rishabh Ranjan, Saurabh Garg, Zachary Lipton","cross_cats":["cs.AI","stat.ML"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-04-11T14:58:19Z","title":"Post-Hoc Reversal: Are We Selecting Models Prematurely?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.07815","kind":"arxiv","version":2},"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":"sha256:872b93bbb867fcada36e64334ed9b9389a967e05c6a85985b5e6acfe59e38d6f","target":"record","created_at":"2026-07-05T09:15:31Z","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":"8b678b19412d031a529332ccbadfdeb34f4b8d13f3364b6fe97cb29767b55f85","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-04-11T14:58:19Z","title_canon_sha256":"66abc60555658b80f0feeca70a874c979f1d70d7361dcce6e486f025e728314d"},"schema_version":"1.0","source":{"id":"2404.07815","kind":"arxiv","version":2}},"canonical_sha256":"483380749b2e5a2babd175710cb0005d70907822e799bc640d8bd90add861da1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"483380749b2e5a2babd175710cb0005d70907822e799bc640d8bd90add861da1","first_computed_at":"2026-07-05T09:15:31.297654Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:15:31.297654Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"oyH3F3E1mxcXdv1EJR1jRvqgmftSIOMb8wj2/Kk9Gkf/zBtOvPTbWmzO4fW/V5GMRpUxxgvsT2ADn/pcJQD3AA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:15:31.298140Z","signed_message":"canonical_sha256_bytes"},"source_id":"2404.07815","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:872b93bbb867fcada36e64334ed9b9389a967e05c6a85985b5e6acfe59e38d6f","sha256:daf1d306910da6a95126538eb989591a313e555982f0db33430024243a6e213e"],"state_sha256":"a6d1be33a14eda8c51d75dafba3c81d2c2a9150d98bb10f3d0384d6b41e1f425"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"B/caQV8tSSOdiCMhP7DtvXZ5Rqh659ACtdCtrNyswj/Js9/OzU/aj6saqkDfZngmuSuxztW5fV4F8Th8Z1dsCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T08:53:18.257397Z","bundle_sha256":"e0e9646188c3761f738cc839500ef0ba012da7ae88cd255043b4dff07ab61c48"}}