{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:KNDH7TRKI76HE57VLZPCIIXCAC","short_pith_number":"pith:KNDH7TRK","canonical_record":{"source":{"id":"2402.00433","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-01T08:58:57Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"089791d68f848cc073a5a7518a50d6a3f00f995491517492bf05b1c83bdb88f5","abstract_canon_sha256":"67d14fc358376adc2f4913ee9eb4e970016ca77942160930ee88c88a3b80a592"},"schema_version":"1.0"},"canonical_sha256":"53467fce2a47fc7277f55e5e2422e200b2f231092d0f3bd7cc1d12dff6e5fc80","source":{"kind":"arxiv","id":"2402.00433","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.00433","created_at":"2026-07-05T08:28:37Z"},{"alias_kind":"arxiv_version","alias_value":"2402.00433v2","created_at":"2026-07-05T08:28:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.00433","created_at":"2026-07-05T08:28:37Z"},{"alias_kind":"pith_short_12","alias_value":"KNDH7TRKI76H","created_at":"2026-07-05T08:28:37Z"},{"alias_kind":"pith_short_16","alias_value":"KNDH7TRKI76HE57V","created_at":"2026-07-05T08:28:37Z"},{"alias_kind":"pith_short_8","alias_value":"KNDH7TRK","created_at":"2026-07-05T08:28:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:KNDH7TRKI76HE57VLZPCIIXCAC","target":"record","payload":{"canonical_record":{"source":{"id":"2402.00433","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-01T08:58:57Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"089791d68f848cc073a5a7518a50d6a3f00f995491517492bf05b1c83bdb88f5","abstract_canon_sha256":"67d14fc358376adc2f4913ee9eb4e970016ca77942160930ee88c88a3b80a592"},"schema_version":"1.0"},"canonical_sha256":"53467fce2a47fc7277f55e5e2422e200b2f231092d0f3bd7cc1d12dff6e5fc80","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:28:37.525479Z","signature_b64":"cBMSCEWgXDk3HVPnaqeDp0IWDEXJXBpD6ChFjZyyNWvxfDDKhyx3FD4sgTg+oGWJv0GYwRye+Ce6Ti0m3U4WAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"53467fce2a47fc7277f55e5e2422e200b2f231092d0f3bd7cc1d12dff6e5fc80","last_reissued_at":"2026-07-05T08:28:37.525042Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:28:37.525042Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2402.00433","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":"2026-07-05T08:28:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nxf9VEAksv+d1GTJO62wBtHuQNo5oUdBNj+Ib3pgcN6vh1zh8Vxg6rgrH3+ZWik6zCNWquSpyniZVKQC2ZKeCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T21:22:58.485835Z"},"content_sha256":"077c8b87757ee9be58f4ed22a280a68ca0304dbdefe52ec8d9c529d3872829a9","schema_version":"1.0","event_id":"sha256:077c8b87757ee9be58f4ed22a280a68ca0304dbdefe52ec8d9c529d3872829a9"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:KNDH7TRKI76HE57VLZPCIIXCAC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Merging Multi-Task Models via Weight-Ensembling Mixture of Experts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Anke Tang, Dacheng Tao, Lefei Zhang, Li Shen, Nan Yin, Yong Luo","submitted_at":"2024-02-01T08:58:57Z","abstract_excerpt":"Merging various task-specific Transformer-based models trained on different tasks into a single unified model can execute all the tasks concurrently. Previous methods, exemplified by task arithmetic, have been proven to be both effective and scalable. Existing methods have primarily focused on seeking a static optimal solution within the original model parameter space. A notable challenge is mitigating the interference between parameters of different models, which can substantially deteriorate performance. In this paper, we propose to merge most of the parameters while upscaling the MLP of the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.00433","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/2402.00433/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":"2026-07-05T08:28:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"W/YcSCOvxp0CnZebx1PFYsmUqds9l8WYcpjnAMA3ToxZ6XQeSBrDuFj+dkKJbFBCHj4hhqRxR8DTIR0yWMQ2BQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T21:22:58.486214Z"},"content_sha256":"5e243d17b1a8d525678b2cda8bfa76fb3597b064c680bf59510dc7e819674bc0","schema_version":"1.0","event_id":"sha256:5e243d17b1a8d525678b2cda8bfa76fb3597b064c680bf59510dc7e819674bc0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KNDH7TRKI76HE57VLZPCIIXCAC/bundle.json","state_url":"https://pith.science/pith/KNDH7TRKI76HE57VLZPCIIXCAC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KNDH7TRKI76HE57VLZPCIIXCAC/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":"2026-08-04T21:22:58Z","links":{"resolver":"https://pith.science/pith/KNDH7TRKI76HE57VLZPCIIXCAC","bundle":"https://pith.science/pith/KNDH7TRKI76HE57VLZPCIIXCAC/bundle.json","state":"https://pith.science/pith/KNDH7TRKI76HE57VLZPCIIXCAC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KNDH7TRKI76HE57VLZPCIIXCAC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:KNDH7TRKI76HE57VLZPCIIXCAC","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":"67d14fc358376adc2f4913ee9eb4e970016ca77942160930ee88c88a3b80a592","cross_cats_sorted":["cs.CV"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-01T08:58:57Z","title_canon_sha256":"089791d68f848cc073a5a7518a50d6a3f00f995491517492bf05b1c83bdb88f5"},"schema_version":"1.0","source":{"id":"2402.00433","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.00433","created_at":"2026-07-05T08:28:37Z"},{"alias_kind":"arxiv_version","alias_value":"2402.00433v2","created_at":"2026-07-05T08:28:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.00433","created_at":"2026-07-05T08:28:37Z"},{"alias_kind":"pith_short_12","alias_value":"KNDH7TRKI76H","created_at":"2026-07-05T08:28:37Z"},{"alias_kind":"pith_short_16","alias_value":"KNDH7TRKI76HE57V","created_at":"2026-07-05T08:28:37Z"},{"alias_kind":"pith_short_8","alias_value":"KNDH7TRK","created_at":"2026-07-05T08:28:37Z"}],"graph_snapshots":[{"event_id":"sha256:5e243d17b1a8d525678b2cda8bfa76fb3597b064c680bf59510dc7e819674bc0","target":"graph","created_at":"2026-07-05T08:28:37Z","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/2402.00433/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Merging various task-specific Transformer-based models trained on different tasks into a single unified model can execute all the tasks concurrently. Previous methods, exemplified by task arithmetic, have been proven to be both effective and scalable. Existing methods have primarily focused on seeking a static optimal solution within the original model parameter space. A notable challenge is mitigating the interference between parameters of different models, which can substantially deteriorate performance. In this paper, we propose to merge most of the parameters while upscaling the MLP of the","authors_text":"Anke Tang, Dacheng Tao, Lefei Zhang, Li Shen, Nan Yin, Yong Luo","cross_cats":["cs.CV"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-01T08:58:57Z","title":"Merging Multi-Task Models via Weight-Ensembling Mixture of Experts"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.00433","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":"sha256:077c8b87757ee9be58f4ed22a280a68ca0304dbdefe52ec8d9c529d3872829a9","target":"record","created_at":"2026-07-05T08:28:37Z","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":"67d14fc358376adc2f4913ee9eb4e970016ca77942160930ee88c88a3b80a592","cross_cats_sorted":["cs.CV"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-01T08:58:57Z","title_canon_sha256":"089791d68f848cc073a5a7518a50d6a3f00f995491517492bf05b1c83bdb88f5"},"schema_version":"1.0","source":{"id":"2402.00433","kind":"arxiv","version":2}},"canonical_sha256":"53467fce2a47fc7277f55e5e2422e200b2f231092d0f3bd7cc1d12dff6e5fc80","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"53467fce2a47fc7277f55e5e2422e200b2f231092d0f3bd7cc1d12dff6e5fc80","first_computed_at":"2026-07-05T08:28:37.525042Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:28:37.525042Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"cBMSCEWgXDk3HVPnaqeDp0IWDEXJXBpD6ChFjZyyNWvxfDDKhyx3FD4sgTg+oGWJv0GYwRye+Ce6Ti0m3U4WAg==","signature_status":"signed_v1","signed_at":"2026-07-05T08:28:37.525479Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.00433","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:077c8b87757ee9be58f4ed22a280a68ca0304dbdefe52ec8d9c529d3872829a9","sha256:5e243d17b1a8d525678b2cda8bfa76fb3597b064c680bf59510dc7e819674bc0"],"state_sha256":"8557f0682f09303c5240b2c7e4a425d5fb7b8c3a912729b5fa63c530d2e5fe51"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rc8P2QOYlU9dF1U/lOMLVUjRWCZysbK+/tg/+D65GVJNNW2vgeWtq+nHvrCGWgN+Xsbnl1plqq+4TmUQndo9AA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T21:22:58.488680Z","bundle_sha256":"3d5778d3908db8b3cf7e588b67228c675431281015fd8f2b5edfd3898d6abbea"}}