{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:RBYLWJVRXNIVOXSNNRZKWEMHHZ","short_pith_number":"pith:RBYLWJVR","schema_version":"1.0","canonical_sha256":"8870bb26b1bb51575e4d6c72ab11873e57e67c490bd1fe092538555dad820f51","source":{"kind":"arxiv","id":"2311.08692","version":1},"attestation_state":"computed","paper":{"title":"Routing to the Expert: Efficient Reward-guided Ensemble of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Chang Zhou, Hongyi Yuan, Jingren Zhou, Junyang Lin, Keming Lu, Runji Lin, Zheng Yuan","submitted_at":"2023-11-15T04:40:43Z","abstract_excerpt":"The complementary potential of Large Language Models (LLM) assumes off-the-shelf LLMs have heterogeneous expertise in a wide range of domains and tasks so that an ensemble of LLMs can achieve consistently better performance. Existing ensemble methods for LLMs mainly focus on reward model ranking of outputs, leading to significant computation overhead. To combat this issue, we revisit the complementary potential of LLMs and further elaborate it by mining latent expertise with off-the-shelf reward models. We propose Zooter, a reward-guided routing method distilling rewards on training queries to"},"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":"2311.08692","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-11-15T04:40:43Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"d909e732479c6b3bff8b2e1577187d3352128eb592d050b7bf58fe535e4d6d0e","abstract_canon_sha256":"f8ceb5b8c2bb660a689126a50f0b09482d4d1b012d5892c60e64d870c8919447"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:13:01.810345Z","signature_b64":"tOafStb6kUyOSyqiMbIjLHjZluQ/Ce74p5cBi/P56juMn7Fso6ptNw17FLaiACNz+DnrIYLjgVod/SaLYx8BCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8870bb26b1bb51575e4d6c72ab11873e57e67c490bd1fe092538555dad820f51","last_reissued_at":"2026-07-05T07:13:01.809847Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:13:01.809847Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Routing to the Expert: Efficient Reward-guided Ensemble of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Chang Zhou, Hongyi Yuan, Jingren Zhou, Junyang Lin, Keming Lu, Runji Lin, Zheng Yuan","submitted_at":"2023-11-15T04:40:43Z","abstract_excerpt":"The complementary potential of Large Language Models (LLM) assumes off-the-shelf LLMs have heterogeneous expertise in a wide range of domains and tasks so that an ensemble of LLMs can achieve consistently better performance. Existing ensemble methods for LLMs mainly focus on reward model ranking of outputs, leading to significant computation overhead. To combat this issue, we revisit the complementary potential of LLMs and further elaborate it by mining latent expertise with off-the-shelf reward models. We propose Zooter, a reward-guided routing method distilling rewards on training queries to"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.08692","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/2311.08692/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":"2311.08692","created_at":"2026-07-05T07:13:01.809909+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.08692v1","created_at":"2026-07-05T07:13:01.809909+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.08692","created_at":"2026-07-05T07:13:01.809909+00:00"},{"alias_kind":"pith_short_12","alias_value":"RBYLWJVRXNIV","created_at":"2026-07-05T07:13:01.809909+00:00"},{"alias_kind":"pith_short_16","alias_value":"RBYLWJVRXNIVOXSN","created_at":"2026-07-05T07:13:01.809909+00:00"},{"alias_kind":"pith_short_8","alias_value":"RBYLWJVR","created_at":"2026-07-05T07:13:01.809909+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.07587","citing_title":"The Routing Plateau: Understanding and Breaking the Accuracy Limits of LLM Routers","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2505.12601","citing_title":"Rethinking Predictive Modeling for LLM Routing: When Simple kNN Beats Complex Learned Routers","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2507.14200","citing_title":"A Scalable Multi-LLM Collaboration System with Retrieval-based Selection and Exploration-Exploitation-Driven Enhancement","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17106","citing_title":"HyDRA: Hybrid Dynamic Routing Architecture for Heterogeneous LLM Pools","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2406.18665","citing_title":"RouteLLM: Learning to Route LLMs with Preference Data","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07395","citing_title":"Unsolvability Ceiling in Multi-LLM Routing: An Empirical Study of Evaluation Artifacts","ref_index":4,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RBYLWJVRXNIVOXSNNRZKWEMHHZ","json":"https://pith.science/pith/RBYLWJVRXNIVOXSNNRZKWEMHHZ.json","graph_json":"https://pith.science/api/pith-number/RBYLWJVRXNIVOXSNNRZKWEMHHZ/graph.json","events_json":"https://pith.science/api/pith-number/RBYLWJVRXNIVOXSNNRZKWEMHHZ/events.json","paper":"https://pith.science/paper/RBYLWJVR"},"agent_actions":{"view_html":"https://pith.science/pith/RBYLWJVRXNIVOXSNNRZKWEMHHZ","download_json":"https://pith.science/pith/RBYLWJVRXNIVOXSNNRZKWEMHHZ.json","view_paper":"https://pith.science/paper/RBYLWJVR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.08692&json=true","fetch_graph":"https://pith.science/api/pith-number/RBYLWJVRXNIVOXSNNRZKWEMHHZ/graph.json","fetch_events":"https://pith.science/api/pith-number/RBYLWJVRXNIVOXSNNRZKWEMHHZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RBYLWJVRXNIVOXSNNRZKWEMHHZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RBYLWJVRXNIVOXSNNRZKWEMHHZ/action/storage_attestation","attest_author":"https://pith.science/pith/RBYLWJVRXNIVOXSNNRZKWEMHHZ/action/author_attestation","sign_citation":"https://pith.science/pith/RBYLWJVRXNIVOXSNNRZKWEMHHZ/action/citation_signature","submit_replication":"https://pith.science/pith/RBYLWJVRXNIVOXSNNRZKWEMHHZ/action/replication_record"}},"created_at":"2026-07-05T07:13:01.809909+00:00","updated_at":"2026-07-05T07:13:01.809909+00:00"}