{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:JV7LWPYNWCBA7VWZRNJHBU6M26","short_pith_number":"pith:JV7LWPYN","schema_version":"1.0","canonical_sha256":"4d7ebb3f0db0820fd6d98b5270d3ccd7af6db927c1ebea92b12fa58a6fd911ec","source":{"kind":"arxiv","id":"2508.10016","version":4},"attestation_state":"computed","paper":{"title":"Training-Free Multimodal Large Language Model Orchestration","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"A training-free framework uses an off-the-shelf LLM to route and sequence separate modality experts into one unified multimodal system.","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jiayi Ji, Rongrong Ji, Tat-Seng Chua, Tianyu Xie, Wang Chen, Xiawu Zheng, Yuexiao Ma, Yuhang Wu","submitted_at":"2025-08-06T16:17:29Z","abstract_excerpt":"Building interactive omni-modal assistants often relies on end-to-end multimodal alignment to fuse heterogeneous modalities, which incurs substantial data and compute costs and limits extensibility. We present Training-Free Large Language Model Orchestration (LLM Orchestration), a training-free orchestration framework that integrates off-the-shelf modality experts into a unified multimodal input--output system without additional gradient-based training for integration. LLM Orchestration comprises three components: (1) an LLM controller that infers user intent and emits explicit control tokens "},"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":true},"canonical_record":{"source":{"id":"2508.10016","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-08-06T16:17:29Z","cross_cats_sorted":[],"title_canon_sha256":"7b2aac3bc4eb52a396b6d73398481ac89f1b6a76341c177dde3603ca54acbc56","abstract_canon_sha256":"42a0e176c9c96a2b7d548a54fa861cb18adb177f81cd8b8323972b2f17495e9a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-25T02:02:08.332537Z","signature_b64":"kozyqsUO1XZL0ezVY32+1A1C4878c3ZgdWhglSdNWP447CwvzDRvjgNp1oo1NzzvqES5amEVdiom0al9SiejBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4d7ebb3f0db0820fd6d98b5270d3ccd7af6db927c1ebea92b12fa58a6fd911ec","last_reissued_at":"2026-05-25T02:02:08.331790Z","signature_status":"signed_v1","first_computed_at":"2026-05-25T02:02:08.331790Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Training-Free Multimodal Large Language Model Orchestration","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"A training-free framework uses an off-the-shelf LLM to route and sequence separate modality experts into one unified multimodal system.","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jiayi Ji, Rongrong Ji, Tat-Seng Chua, Tianyu Xie, Wang Chen, Xiawu Zheng, Yuexiao Ma, Yuhang Wu","submitted_at":"2025-08-06T16:17:29Z","abstract_excerpt":"Building interactive omni-modal assistants often relies on end-to-end multimodal alignment to fuse heterogeneous modalities, which incurs substantial data and compute costs and limits extensibility. We present Training-Free Large Language Model Orchestration (LLM Orchestration), a training-free orchestration framework that integrates off-the-shelf modality experts into a unified multimodal input--output system without additional gradient-based training for integration. LLM Orchestration comprises three components: (1) an LLM controller that infers user intent and emits explicit control tokens "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Across diverse multimodal benchmarks, LLM Orchestration achieves strong performance under standard evaluation constraints while maintaining low orchestration overhead and modular upgradeability, providing a practical alternative to costly joint training for omni-modal systems.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That an off-the-shelf LLM controller can reliably infer user intent and emit correct explicit control tokens for expert selection and sequencing without introducing routing errors that degrade overall performance.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A training-free orchestration framework integrates off-the-shelf modality experts via an LLM controller, text-centric cross-modal memory, and unified interaction layer to enable multimodal input-output without joint training.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A training-free framework uses an off-the-shelf LLM to route and sequence separate modality experts into one unified multimodal system.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"bc00d38d97cb8b59067d36e045c46db0905d74748539a2eb1537c9ec35cd39d1"},"source":{"id":"2508.10016","kind":"arxiv","version":4},"verdict":{"id":"d5843d97-2344-442c-b6f0-8178b8168352","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T00:09:26.309979Z","strongest_claim":"Across diverse multimodal benchmarks, LLM Orchestration achieves strong performance under standard evaluation constraints while maintaining low orchestration overhead and modular upgradeability, providing a practical alternative to costly joint training for omni-modal systems.","one_line_summary":"A training-free orchestration framework integrates off-the-shelf modality experts via an LLM controller, text-centric cross-modal memory, and unified interaction layer to enable multimodal input-output without joint training.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That an off-the-shelf LLM controller can reliably infer user intent and emit correct explicit control tokens for expert selection and sequencing without introducing routing errors that degrade overall performance.","pith_extraction_headline":"A training-free framework uses an off-the-shelf LLM to route and sequence separate modality experts into one unified multimodal system."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2508.10016/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":2,"snapshot_sha256":"c8f3035a31a58b1b80f66c0530d5838a8af6aca5b6948db126fd44fdb7b8e73c"},"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":"2508.10016","created_at":"2026-05-25T02:02:08.331871+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.10016v4","created_at":"2026-05-25T02:02:08.331871+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.10016","created_at":"2026-05-25T02:02:08.331871+00:00"},{"alias_kind":"pith_short_12","alias_value":"JV7LWPYNWCBA","created_at":"2026-05-25T02:02:08.331871+00:00"},{"alias_kind":"pith_short_16","alias_value":"JV7LWPYNWCBA7VWZ","created_at":"2026-05-25T02:02:08.331871+00:00"},{"alias_kind":"pith_short_8","alias_value":"JV7LWPYN","created_at":"2026-05-25T02:02:08.331871+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JV7LWPYNWCBA7VWZRNJHBU6M26","json":"https://pith.science/pith/JV7LWPYNWCBA7VWZRNJHBU6M26.json","graph_json":"https://pith.science/api/pith-number/JV7LWPYNWCBA7VWZRNJHBU6M26/graph.json","events_json":"https://pith.science/api/pith-number/JV7LWPYNWCBA7VWZRNJHBU6M26/events.json","paper":"https://pith.science/paper/JV7LWPYN"},"agent_actions":{"view_html":"https://pith.science/pith/JV7LWPYNWCBA7VWZRNJHBU6M26","download_json":"https://pith.science/pith/JV7LWPYNWCBA7VWZRNJHBU6M26.json","view_paper":"https://pith.science/paper/JV7LWPYN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.10016&json=true","fetch_graph":"https://pith.science/api/pith-number/JV7LWPYNWCBA7VWZRNJHBU6M26/graph.json","fetch_events":"https://pith.science/api/pith-number/JV7LWPYNWCBA7VWZRNJHBU6M26/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JV7LWPYNWCBA7VWZRNJHBU6M26/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JV7LWPYNWCBA7VWZRNJHBU6M26/action/storage_attestation","attest_author":"https://pith.science/pith/JV7LWPYNWCBA7VWZRNJHBU6M26/action/author_attestation","sign_citation":"https://pith.science/pith/JV7LWPYNWCBA7VWZRNJHBU6M26/action/citation_signature","submit_replication":"https://pith.science/pith/JV7LWPYNWCBA7VWZRNJHBU6M26/action/replication_record"}},"created_at":"2026-05-25T02:02:08.331871+00:00","updated_at":"2026-05-25T02:02:08.331871+00:00"}