{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:MBWEQDKASRJBFPVLGGZMB6NXP4","short_pith_number":"pith:MBWEQDKA","schema_version":"1.0","canonical_sha256":"606c480d40945212beab31b2c0f9b77f37c7ab74f1b911b51620bbe2b9973bae","source":{"kind":"arxiv","id":"2605.18710","version":1},"attestation_state":"computed","paper":{"title":"Mosaic: Towards Efficient Training of Multimodal Models with Spatial Resource Multiplexing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Anbang Wu, Chen Chen, Chunyu Xue, Qizhen Weng, Quan Chen, Yanbo Wang, Yin Chen, Yu Feng, Yuxuan Wang","submitted_at":"2026-05-18T17:44:29Z","abstract_excerpt":"With the wide adoption of Multimodal Models (MMs) in real-world scenarios, it is significant to efficiently train emerging MMs that exhibit increasingly complex module architectures. For MM deployment, existing works allocate a GPU to only one MM module in a temporal-multiplexing manner; this compromises training efficiency because a single module often fails to achieve high GPU utilization. To improve GPU utilization and enable efficient MM training, we propose deploying MMs in a temporal-spatial multiplexing manner, allowing multiple MM modules to colocate on a GPU with well-controlled resou"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2605.18710","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2026-05-18T17:44:29Z","cross_cats_sorted":[],"title_canon_sha256":"eae616d4169f48c0336242ff691e57d5c83aca1ee406c23f67cfc92815127476","abstract_canon_sha256":"138f456466dda2580c15859415128c3e4ebd8dc2bc6550a14f030d4fae58eaa3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:06:16.487076Z","signature_b64":"/yCc/P8iKPmem55DcmszAdJeYQd2+/D/cCkOV4Mb6U4tbPsssrqgnZT11rmZltaO/DlCE5JWxODJXNK4XmkwAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"606c480d40945212beab31b2c0f9b77f37c7ab74f1b911b51620bbe2b9973bae","last_reissued_at":"2026-05-20T00:06:16.486346Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:06:16.486346Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Mosaic: Towards Efficient Training of Multimodal Models with Spatial Resource Multiplexing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Anbang Wu, Chen Chen, Chunyu Xue, Qizhen Weng, Quan Chen, Yanbo Wang, Yin Chen, Yu Feng, Yuxuan Wang","submitted_at":"2026-05-18T17:44:29Z","abstract_excerpt":"With the wide adoption of Multimodal Models (MMs) in real-world scenarios, it is significant to efficiently train emerging MMs that exhibit increasingly complex module architectures. For MM deployment, existing works allocate a GPU to only one MM module in a temporal-multiplexing manner; this compromises training efficiency because a single module often fails to achieve high GPU utilization. To improve GPU utilization and enable efficient MM training, we propose deploying MMs in a temporal-spatial multiplexing manner, allowing multiple MM modules to colocate on a GPU with well-controlled resou"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.18710","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/2605.18710/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-20T00:01:59.060021Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"a0a25b0fed78f5faf5296dbca98426f6fa7e67217c86ca5f44edcefa1d9d9b4a"},"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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2605.18710","created_at":"2026-05-20T00:06:16.486458+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.18710v1","created_at":"2026-05-20T00:06:16.486458+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.18710","created_at":"2026-05-20T00:06:16.486458+00:00"},{"alias_kind":"pith_short_12","alias_value":"MBWEQDKASRJB","created_at":"2026-05-20T00:06:16.486458+00:00"},{"alias_kind":"pith_short_16","alias_value":"MBWEQDKASRJBFPVL","created_at":"2026-05-20T00:06:16.486458+00:00"},{"alias_kind":"pith_short_8","alias_value":"MBWEQDKA","created_at":"2026-05-20T00:06:16.486458+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MBWEQDKASRJBFPVLGGZMB6NXP4","json":"https://pith.science/pith/MBWEQDKASRJBFPVLGGZMB6NXP4.json","graph_json":"https://pith.science/api/pith-number/MBWEQDKASRJBFPVLGGZMB6NXP4/graph.json","events_json":"https://pith.science/api/pith-number/MBWEQDKASRJBFPVLGGZMB6NXP4/events.json","paper":"https://pith.science/paper/MBWEQDKA"},"agent_actions":{"view_html":"https://pith.science/pith/MBWEQDKASRJBFPVLGGZMB6NXP4","download_json":"https://pith.science/pith/MBWEQDKASRJBFPVLGGZMB6NXP4.json","view_paper":"https://pith.science/paper/MBWEQDKA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.18710&json=true","fetch_graph":"https://pith.science/api/pith-number/MBWEQDKASRJBFPVLGGZMB6NXP4/graph.json","fetch_events":"https://pith.science/api/pith-number/MBWEQDKASRJBFPVLGGZMB6NXP4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MBWEQDKASRJBFPVLGGZMB6NXP4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MBWEQDKASRJBFPVLGGZMB6NXP4/action/storage_attestation","attest_author":"https://pith.science/pith/MBWEQDKASRJBFPVLGGZMB6NXP4/action/author_attestation","sign_citation":"https://pith.science/pith/MBWEQDKASRJBFPVLGGZMB6NXP4/action/citation_signature","submit_replication":"https://pith.science/pith/MBWEQDKASRJBFPVLGGZMB6NXP4/action/replication_record"}},"created_at":"2026-05-20T00:06:16.486458+00:00","updated_at":"2026-05-20T00:06:16.486458+00:00"}