{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:N33MRVA5DZKI2D6GXKT662C5ZF","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":"3f3965f761e57ae9ca841af054f0d3717a90a3fef91ca52e95e4f8b9ea373706","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-03T20:33:20Z","title_canon_sha256":"cf0cc282781c4741f40664cc2ed4f703264ebaf3e31b70a5142286ac5a67e41e"},"schema_version":"1.0","source":{"id":"2502.01804","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.01804","created_at":"2026-07-05T10:09:19Z"},{"alias_kind":"arxiv_version","alias_value":"2502.01804v1","created_at":"2026-07-05T10:09:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.01804","created_at":"2026-07-05T10:09:19Z"},{"alias_kind":"pith_short_12","alias_value":"N33MRVA5DZKI","created_at":"2026-07-05T10:09:19Z"},{"alias_kind":"pith_short_16","alias_value":"N33MRVA5DZKI2D6G","created_at":"2026-07-05T10:09:19Z"},{"alias_kind":"pith_short_8","alias_value":"N33MRVA5","created_at":"2026-07-05T10:09:19Z"}],"graph_snapshots":[{"event_id":"sha256:54b72429185ad3fa91ec8e1aa1ee36e350e7ab67d72b98eef9942cefc754f873","target":"graph","created_at":"2026-07-05T10:09:19Z","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/2502.01804/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Machine learning models are routinely trained on a mixture of different data domains. Different domain weights yield very different downstream performances. We propose the Soup-of-Experts, a novel architecture that can instantiate a model at test time for any domain weights with minimal computational cost and without re-training the model. Our architecture consists of a bank of expert parameters, which are linearly combined to instantiate one model. We learn the linear combination coefficients as a function of the input domain weights. To train this architecture, we sample random domain weight","authors_text":"Angelos Katharopoulos, David Grangier, Pierre Ablin, Skyler Seto","cross_cats":["cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-03T20:33:20Z","title":"Soup-of-Experts: Pretraining Specialist Models via Parameters Averaging"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.01804","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":"sha256:4bc8d6926e95be3a0a5a250ec67f0a8ab9234904aa32c3a3fd88555a5d451f52","target":"record","created_at":"2026-07-05T10:09:19Z","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":"3f3965f761e57ae9ca841af054f0d3717a90a3fef91ca52e95e4f8b9ea373706","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-03T20:33:20Z","title_canon_sha256":"cf0cc282781c4741f40664cc2ed4f703264ebaf3e31b70a5142286ac5a67e41e"},"schema_version":"1.0","source":{"id":"2502.01804","kind":"arxiv","version":1}},"canonical_sha256":"6ef6c8d41d1e548d0fc6baa7ef685dc9626edf2ad8f6c07733bd485568e4bdee","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6ef6c8d41d1e548d0fc6baa7ef685dc9626edf2ad8f6c07733bd485568e4bdee","first_computed_at":"2026-07-05T10:09:19.429750Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:09:19.429750Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"tys0Sa8dKcQK+MQPb+Gq1Yb8AUd+t22ZTgBh0X/3ebY6ymI3BR6OopzWHd+TDk/51CGSTg+cn/O3P+OoK5VrAg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:09:19.430358Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.01804","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4bc8d6926e95be3a0a5a250ec67f0a8ab9234904aa32c3a3fd88555a5d451f52","sha256:54b72429185ad3fa91ec8e1aa1ee36e350e7ab67d72b98eef9942cefc754f873"],"state_sha256":"c7d3b9c688f05c73e435b65a84565f9bdd699ff9a2fef14364f2511822761448"}