{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:VJ2KNKHPOWJX3UYL6US5NR2CGR","short_pith_number":"pith:VJ2KNKHP","schema_version":"1.0","canonical_sha256":"aa74a6a8ef75937dd30bf525d6c742347b4eea723091d0476208e996364bff24","source":{"kind":"arxiv","id":"2402.16107","version":6},"attestation_state":"computed","paper":{"title":"Knowledge Fusion of Chat LLMs: A Preliminary Technical Report","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Fanqi Wan, Longguang Zhong, Wei Bi, Xiaojun Quan, Xinting Huang, Ziyi Yang","submitted_at":"2024-02-25T15:11:58Z","abstract_excerpt":"Recently, FuseLLM introduced the concept of knowledge fusion to transfer the collective knowledge of multiple structurally varied LLMs into a target LLM through lightweight continual training. In this report, we extend the scalability and flexibility of the FuseLLM framework to realize the fusion of chat LLMs, resulting in FusionChat. FusionChat comprises two main stages. Firstly, we undertake knowledge fusion for structurally and scale-varied source LLMs to derive multiple target LLMs of identical structure and size via lightweight fine-tuning. Then, these target LLMs are merged within the pa"},"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":"2402.16107","kind":"arxiv","version":6},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-25T15:11:58Z","cross_cats_sorted":[],"title_canon_sha256":"51c415e9246487b2f4165b16ce6aea82b42b65b34e45478be96df29e4c3a6664","abstract_canon_sha256":"3671cf4a0d0b55f897edd8358656797bc6de68f260b14734940f1e64ba1257c2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:26:43.857904Z","signature_b64":"UkhfStyoHw5whL21nV/oh0Pj3cQpyVtmKT29BDChUgFJhH1tSjai8z8qN8DoxTOCXNx1MmC7Y37iLqQUo7ZnAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"aa74a6a8ef75937dd30bf525d6c742347b4eea723091d0476208e996364bff24","last_reissued_at":"2026-07-05T08:26:43.857434Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:26:43.857434Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Knowledge Fusion of Chat LLMs: A Preliminary Technical Report","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Fanqi Wan, Longguang Zhong, Wei Bi, Xiaojun Quan, Xinting Huang, Ziyi Yang","submitted_at":"2024-02-25T15:11:58Z","abstract_excerpt":"Recently, FuseLLM introduced the concept of knowledge fusion to transfer the collective knowledge of multiple structurally varied LLMs into a target LLM through lightweight continual training. In this report, we extend the scalability and flexibility of the FuseLLM framework to realize the fusion of chat LLMs, resulting in FusionChat. FusionChat comprises two main stages. Firstly, we undertake knowledge fusion for structurally and scale-varied source LLMs to derive multiple target LLMs of identical structure and size via lightweight fine-tuning. Then, these target LLMs are merged within the pa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.16107","kind":"arxiv","version":6},"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/2402.16107/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2402.16107","created_at":"2026-07-05T08:26:43.857491+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.16107v6","created_at":"2026-07-05T08:26:43.857491+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.16107","created_at":"2026-07-05T08:26:43.857491+00:00"},{"alias_kind":"pith_short_12","alias_value":"VJ2KNKHPOWJX","created_at":"2026-07-05T08:26:43.857491+00:00"},{"alias_kind":"pith_short_16","alias_value":"VJ2KNKHPOWJX3UYL","created_at":"2026-07-05T08:26:43.857491+00:00"},{"alias_kind":"pith_short_8","alias_value":"VJ2KNKHP","created_at":"2026-07-05T08:26:43.857491+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2408.07666","citing_title":"Model Merging in LLMs, MLLMs, and Beyond: Methods, Theories, Applications and Opportunities","ref_index":233,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VJ2KNKHPOWJX3UYL6US5NR2CGR","json":"https://pith.science/pith/VJ2KNKHPOWJX3UYL6US5NR2CGR.json","graph_json":"https://pith.science/api/pith-number/VJ2KNKHPOWJX3UYL6US5NR2CGR/graph.json","events_json":"https://pith.science/api/pith-number/VJ2KNKHPOWJX3UYL6US5NR2CGR/events.json","paper":"https://pith.science/paper/VJ2KNKHP"},"agent_actions":{"view_html":"https://pith.science/pith/VJ2KNKHPOWJX3UYL6US5NR2CGR","download_json":"https://pith.science/pith/VJ2KNKHPOWJX3UYL6US5NR2CGR.json","view_paper":"https://pith.science/paper/VJ2KNKHP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.16107&json=true","fetch_graph":"https://pith.science/api/pith-number/VJ2KNKHPOWJX3UYL6US5NR2CGR/graph.json","fetch_events":"https://pith.science/api/pith-number/VJ2KNKHPOWJX3UYL6US5NR2CGR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VJ2KNKHPOWJX3UYL6US5NR2CGR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VJ2KNKHPOWJX3UYL6US5NR2CGR/action/storage_attestation","attest_author":"https://pith.science/pith/VJ2KNKHPOWJX3UYL6US5NR2CGR/action/author_attestation","sign_citation":"https://pith.science/pith/VJ2KNKHPOWJX3UYL6US5NR2CGR/action/citation_signature","submit_replication":"https://pith.science/pith/VJ2KNKHPOWJX3UYL6US5NR2CGR/action/replication_record"}},"created_at":"2026-07-05T08:26:43.857491+00:00","updated_at":"2026-07-05T08:26:43.857491+00:00"}