{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:XSNBFLXXVUK7HHTF27K2TZ7B5K","short_pith_number":"pith:XSNBFLXX","canonical_record":{"source":{"id":"2405.17517","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-05-27T09:02:57Z","cross_cats_sorted":["cs.CV","cs.NE","stat.ML"],"title_canon_sha256":"98ab9c4d579853a96cee84663dafa7dfac1a5c13305933cc0b502556cb05c41e","abstract_canon_sha256":"2314e055fe4fe103aad228e8893e67ae8d0cb4760352e54cef187561315b0703"},"schema_version":"1.0"},"canonical_sha256":"bc9a12aef7ad15f39e65d7d5a9e7e1ea80361a1c7be41762844ac2bdd6348e34","source":{"kind":"arxiv","id":"2405.17517","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.17517","created_at":"2026-07-05T08:24:07Z"},{"alias_kind":"arxiv_version","alias_value":"2405.17517v1","created_at":"2026-07-05T08:24:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.17517","created_at":"2026-07-05T08:24:07Z"},{"alias_kind":"pith_short_12","alias_value":"XSNBFLXXVUK7","created_at":"2026-07-05T08:24:07Z"},{"alias_kind":"pith_short_16","alias_value":"XSNBFLXXVUK7HHTF","created_at":"2026-07-05T08:24:07Z"},{"alias_kind":"pith_short_8","alias_value":"XSNBFLXX","created_at":"2026-07-05T08:24:07Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:XSNBFLXXVUK7HHTF27K2TZ7B5K","target":"record","payload":{"canonical_record":{"source":{"id":"2405.17517","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-05-27T09:02:57Z","cross_cats_sorted":["cs.CV","cs.NE","stat.ML"],"title_canon_sha256":"98ab9c4d579853a96cee84663dafa7dfac1a5c13305933cc0b502556cb05c41e","abstract_canon_sha256":"2314e055fe4fe103aad228e8893e67ae8d0cb4760352e54cef187561315b0703"},"schema_version":"1.0"},"canonical_sha256":"bc9a12aef7ad15f39e65d7d5a9e7e1ea80361a1c7be41762844ac2bdd6348e34","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:24:07.904995Z","signature_b64":"DbE6S+bbURtE7ZJ4EA/DUBWxoRAWgVcoZW+kYDvQDmDpWLH+a43b9r85MJPYsVoRbvtmBIWEj9xh4/IGs469DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bc9a12aef7ad15f39e65d7d5a9e7e1ea80361a1c7be41762844ac2bdd6348e34","last_reissued_at":"2026-07-05T08:24:07.904571Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:24:07.904571Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2405.17517","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":"2026-07-05T08:24:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JmZKoFHocddPdhm1ORrfOWvM2JhSAlYhFvOyQL9icOkQtgcADkZ34HBJUROiSxyjVsZBFXo/EnkJ6uVhT7mBDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T16:23:32.963778Z"},"content_sha256":"d459f5e5e6104e0257b593de349c5250d024a5f88046f47cc08fdd90a66c74dc","schema_version":"1.0","event_id":"sha256:d459f5e5e6104e0257b593de349c5250d024a5f88046f47cc08fdd90a66c74dc"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:XSNBFLXXVUK7HHTF27K2TZ7B5K","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"WASH: Train your Ensemble with Communication-Efficient Weight Shuffling, then Average","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","cs.NE","stat.ML"],"primary_cat":"cs.LG","authors_text":"Adel Nabli (MLIA, Edouard Oyallon, Eugene Belilovsky (MILA), Louis Fournier (MLIA), Marco Pedersoli (ETS), Masih Aminbeidokhti (ETS), MILA)","submitted_at":"2024-05-27T09:02:57Z","abstract_excerpt":"The performance of deep neural networks is enhanced by ensemble methods, which average the output of several models. However, this comes at an increased cost at inference. Weight averaging methods aim at balancing the generalization of ensembling and the inference speed of a single model by averaging the parameters of an ensemble of models. Yet, naive averaging results in poor performance as models converge to different loss basins, and aligning the models to improve the performance of the average is challenging. Alternatively, inspired by distributed training, methods like DART and PAPA have "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.17517","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/2405.17517/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":"2026-07-05T08:24:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"inGt8YKILrjPGYbBZvG+Yv+PGhDKsXKWB3t04UN69++dH7aQhVqYAO3qkkp4M06fHDDHI3LSZl50vFHMYFpOAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T16:23:32.964290Z"},"content_sha256":"482d40e0ae09ea68599ad78a989b2d1ae975d39f2ce905cbf74ba2a362d15b9d","schema_version":"1.0","event_id":"sha256:482d40e0ae09ea68599ad78a989b2d1ae975d39f2ce905cbf74ba2a362d15b9d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XSNBFLXXVUK7HHTF27K2TZ7B5K/bundle.json","state_url":"https://pith.science/pith/XSNBFLXXVUK7HHTF27K2TZ7B5K/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XSNBFLXXVUK7HHTF27K2TZ7B5K/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":"2026-08-08T16:23:32Z","links":{"resolver":"https://pith.science/pith/XSNBFLXXVUK7HHTF27K2TZ7B5K","bundle":"https://pith.science/pith/XSNBFLXXVUK7HHTF27K2TZ7B5K/bundle.json","state":"https://pith.science/pith/XSNBFLXXVUK7HHTF27K2TZ7B5K/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XSNBFLXXVUK7HHTF27K2TZ7B5K/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:XSNBFLXXVUK7HHTF27K2TZ7B5K","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":"2314e055fe4fe103aad228e8893e67ae8d0cb4760352e54cef187561315b0703","cross_cats_sorted":["cs.CV","cs.NE","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-05-27T09:02:57Z","title_canon_sha256":"98ab9c4d579853a96cee84663dafa7dfac1a5c13305933cc0b502556cb05c41e"},"schema_version":"1.0","source":{"id":"2405.17517","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.17517","created_at":"2026-07-05T08:24:07Z"},{"alias_kind":"arxiv_version","alias_value":"2405.17517v1","created_at":"2026-07-05T08:24:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.17517","created_at":"2026-07-05T08:24:07Z"},{"alias_kind":"pith_short_12","alias_value":"XSNBFLXXVUK7","created_at":"2026-07-05T08:24:07Z"},{"alias_kind":"pith_short_16","alias_value":"XSNBFLXXVUK7HHTF","created_at":"2026-07-05T08:24:07Z"},{"alias_kind":"pith_short_8","alias_value":"XSNBFLXX","created_at":"2026-07-05T08:24:07Z"}],"graph_snapshots":[{"event_id":"sha256:482d40e0ae09ea68599ad78a989b2d1ae975d39f2ce905cbf74ba2a362d15b9d","target":"graph","created_at":"2026-07-05T08:24:07Z","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/2405.17517/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The performance of deep neural networks is enhanced by ensemble methods, which average the output of several models. However, this comes at an increased cost at inference. Weight averaging methods aim at balancing the generalization of ensembling and the inference speed of a single model by averaging the parameters of an ensemble of models. Yet, naive averaging results in poor performance as models converge to different loss basins, and aligning the models to improve the performance of the average is challenging. Alternatively, inspired by distributed training, methods like DART and PAPA have ","authors_text":"Adel Nabli (MLIA, Edouard Oyallon, Eugene Belilovsky (MILA), Louis Fournier (MLIA), Marco Pedersoli (ETS), Masih Aminbeidokhti (ETS), MILA)","cross_cats":["cs.CV","cs.NE","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-05-27T09:02:57Z","title":"WASH: Train your Ensemble with Communication-Efficient Weight Shuffling, then Average"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.17517","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:d459f5e5e6104e0257b593de349c5250d024a5f88046f47cc08fdd90a66c74dc","target":"record","created_at":"2026-07-05T08:24:07Z","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":"2314e055fe4fe103aad228e8893e67ae8d0cb4760352e54cef187561315b0703","cross_cats_sorted":["cs.CV","cs.NE","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-05-27T09:02:57Z","title_canon_sha256":"98ab9c4d579853a96cee84663dafa7dfac1a5c13305933cc0b502556cb05c41e"},"schema_version":"1.0","source":{"id":"2405.17517","kind":"arxiv","version":1}},"canonical_sha256":"bc9a12aef7ad15f39e65d7d5a9e7e1ea80361a1c7be41762844ac2bdd6348e34","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"bc9a12aef7ad15f39e65d7d5a9e7e1ea80361a1c7be41762844ac2bdd6348e34","first_computed_at":"2026-07-05T08:24:07.904571Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:24:07.904571Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"DbE6S+bbURtE7ZJ4EA/DUBWxoRAWgVcoZW+kYDvQDmDpWLH+a43b9r85MJPYsVoRbvtmBIWEj9xh4/IGs469DQ==","signature_status":"signed_v1","signed_at":"2026-07-05T08:24:07.904995Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.17517","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d459f5e5e6104e0257b593de349c5250d024a5f88046f47cc08fdd90a66c74dc","sha256:482d40e0ae09ea68599ad78a989b2d1ae975d39f2ce905cbf74ba2a362d15b9d"],"state_sha256":"bf6c2bc4dca9c8af08e32a55ffafe5cf41c890c682ccefcfbecc3744050d5fcc"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"u4ZhNeVYqofDjs7l6E3Ku3IehmGIHlnbz3Gj4HqFyTWjTx5Ai82dJc/BZC933MT855l1J+xJNHlwE5mxDIZ2Bw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T16:23:32.969345Z","bundle_sha256":"bf1fc6f2d16420f4691812c12cca04fe0a97863d63bd4cb36408a210d8155576"}}