{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:CGQY6QH3G23POE2V2NHDMBPFJL","short_pith_number":"pith:CGQY6QH3","schema_version":"1.0","canonical_sha256":"11a18f40fb36b6f71355d34e3605e54af65ff971b8baf9bbc24641938ca64623","source":{"kind":"arxiv","id":"2410.03953","version":1},"attestation_state":"computed","paper":{"title":"LLM-TOPLA: Efficient LLM Ensemble by Maximising Diversity","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Fatih Ilhan, Ling Liu, Selim Furkan Tekin, Sihao Hu, Tiansheng Huang","submitted_at":"2024-10-04T22:31:15Z","abstract_excerpt":"Combining large language models during training or at inference time has shown substantial performance gain over component LLMs. This paper presents LLM-TOPLA, a diversity-optimized LLM ensemble method with three unique properties: (i) We introduce the focal diversity metric to capture the diversity-performance correlation among component LLMs of an ensemble. (ii) We develop a diversity-optimized ensemble pruning algorithm to select the top-k sub-ensembles from a pool of $N$ base LLMs. Our pruning method recommends top-performing LLM subensembles of size $S$, often much smaller than $N$. (iii)"},"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":"2410.03953","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-04T22:31:15Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"abf4812f600796b12cd778ec2482090fbe1e2c199f170889985a63ae84bc7381","abstract_canon_sha256":"cf0add0ee8a6458a9350a3a1641b953f56c4489d2dab9affd8e532e2ca531062"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:16:26.897246Z","signature_b64":"v5AEEqK2T1cYn8pj6iyafY+SA4IObFzfbplxmBaMlZxv/qK4GCFE3dw4wPYza2zNgtFzik9UWz32sSeV4+pjAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"11a18f40fb36b6f71355d34e3605e54af65ff971b8baf9bbc24641938ca64623","last_reissued_at":"2026-07-05T09:16:26.896770Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:16:26.896770Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LLM-TOPLA: Efficient LLM Ensemble by Maximising Diversity","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Fatih Ilhan, Ling Liu, Selim Furkan Tekin, Sihao Hu, Tiansheng Huang","submitted_at":"2024-10-04T22:31:15Z","abstract_excerpt":"Combining large language models during training or at inference time has shown substantial performance gain over component LLMs. This paper presents LLM-TOPLA, a diversity-optimized LLM ensemble method with three unique properties: (i) We introduce the focal diversity metric to capture the diversity-performance correlation among component LLMs of an ensemble. (ii) We develop a diversity-optimized ensemble pruning algorithm to select the top-k sub-ensembles from a pool of $N$ base LLMs. Our pruning method recommends top-performing LLM subensembles of size $S$, often much smaller than $N$. (iii)"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.03953","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/2410.03953/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":"2410.03953","created_at":"2026-07-05T09:16:26.896831+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.03953v1","created_at":"2026-07-05T09:16:26.896831+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.03953","created_at":"2026-07-05T09:16:26.896831+00:00"},{"alias_kind":"pith_short_12","alias_value":"CGQY6QH3G23P","created_at":"2026-07-05T09:16:26.896831+00:00"},{"alias_kind":"pith_short_16","alias_value":"CGQY6QH3G23POE2V","created_at":"2026-07-05T09:16:26.896831+00:00"},{"alias_kind":"pith_short_8","alias_value":"CGQY6QH3","created_at":"2026-07-05T09:16:26.896831+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/CGQY6QH3G23POE2V2NHDMBPFJL","json":"https://pith.science/pith/CGQY6QH3G23POE2V2NHDMBPFJL.json","graph_json":"https://pith.science/api/pith-number/CGQY6QH3G23POE2V2NHDMBPFJL/graph.json","events_json":"https://pith.science/api/pith-number/CGQY6QH3G23POE2V2NHDMBPFJL/events.json","paper":"https://pith.science/paper/CGQY6QH3"},"agent_actions":{"view_html":"https://pith.science/pith/CGQY6QH3G23POE2V2NHDMBPFJL","download_json":"https://pith.science/pith/CGQY6QH3G23POE2V2NHDMBPFJL.json","view_paper":"https://pith.science/paper/CGQY6QH3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.03953&json=true","fetch_graph":"https://pith.science/api/pith-number/CGQY6QH3G23POE2V2NHDMBPFJL/graph.json","fetch_events":"https://pith.science/api/pith-number/CGQY6QH3G23POE2V2NHDMBPFJL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CGQY6QH3G23POE2V2NHDMBPFJL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CGQY6QH3G23POE2V2NHDMBPFJL/action/storage_attestation","attest_author":"https://pith.science/pith/CGQY6QH3G23POE2V2NHDMBPFJL/action/author_attestation","sign_citation":"https://pith.science/pith/CGQY6QH3G23POE2V2NHDMBPFJL/action/citation_signature","submit_replication":"https://pith.science/pith/CGQY6QH3G23POE2V2NHDMBPFJL/action/replication_record"}},"created_at":"2026-07-05T09:16:26.896831+00:00","updated_at":"2026-07-05T09:16:26.896831+00:00"}