{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:64LSJYUEBNM4XBDFFQ3JYJT2NB","short_pith_number":"pith:64LSJYUE","schema_version":"1.0","canonical_sha256":"f71724e2840b59cb84652c369c267a687507d4916d934bb118df9f35108564d6","source":{"kind":"arxiv","id":"2410.10347","version":3},"attestation_state":"computed","paper":{"title":"A Unified Approach to Routing and Cascading for LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jasper Dekoninck, Martin Vechev, Maximilian Baader","submitted_at":"2024-10-14T10:00:49Z","abstract_excerpt":"The availability of a wide range of large language models (LLMs) embedded in various agentic systems has significantly increased the potential of model selection strategies to improve the cost-performance tradeoff. Existing strategies involve either routing, where a single model is chosen per query, or cascading, which sequentially runs increasingly larger models until a satisfactory answer is found. However, current approaches face three key limitations: they (1) lack formal proofs of optimality, (2) fail to identify the conditions under which these strategies are most effective to improve th"},"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":"2410.10347","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-14T10:00:49Z","cross_cats_sorted":[],"title_canon_sha256":"83a438af189350c6782ec3d78f1b4909b6543e7fbede4abb4b5541db48edaf23","abstract_canon_sha256":"891e75e1676fd9a10be8eaebe89aa710eae4de5749dcec1d4a52b72d04224ee2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:07:15.534297Z","signature_b64":"GGi0Ora50pLtxhVSKd56R3H5DdZ3f4P8DsAH8xWk3RW4038HCI+fAzPUa916Xjf65ym9IoB/e6+PFUfaiickCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f71724e2840b59cb84652c369c267a687507d4916d934bb118df9f35108564d6","last_reissued_at":"2026-07-05T11:07:15.533800Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:07:15.533800Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Unified Approach to Routing and Cascading for LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jasper Dekoninck, Martin Vechev, Maximilian Baader","submitted_at":"2024-10-14T10:00:49Z","abstract_excerpt":"The availability of a wide range of large language models (LLMs) embedded in various agentic systems has significantly increased the potential of model selection strategies to improve the cost-performance tradeoff. Existing strategies involve either routing, where a single model is chosen per query, or cascading, which sequentially runs increasingly larger models until a satisfactory answer is found. However, current approaches face three key limitations: they (1) lack formal proofs of optimality, (2) fail to identify the conditions under which these strategies are most effective to improve th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.10347","kind":"arxiv","version":3},"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/2410.10347/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":"2410.10347","created_at":"2026-07-05T11:07:15.533865+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.10347v3","created_at":"2026-07-05T11:07:15.533865+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.10347","created_at":"2026-07-05T11:07:15.533865+00:00"},{"alias_kind":"pith_short_12","alias_value":"64LSJYUEBNM4","created_at":"2026-07-05T11:07:15.533865+00:00"},{"alias_kind":"pith_short_16","alias_value":"64LSJYUEBNM4XBDF","created_at":"2026-07-05T11:07:15.533865+00:00"},{"alias_kind":"pith_short_8","alias_value":"64LSJYUE","created_at":"2026-07-05T11:07:15.533865+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.25871","citing_title":"AutoRelAnnotator: Calibrated Model Cascades for Cost-Efficient Relevance Evaluation in Sponsored Search","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2606.27288","citing_title":"When Does Combining Language Models Help? A Co-Failure Ceiling on Routing, Voting, and Mixture-of-Agents Across 67 Frontier Models","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2606.20713","citing_title":"FairTutor: Equity-Aware Pedagogical LLM Routing for Budget-Constrained AI Tutoring","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2502.18036","citing_title":"Harnessing Multiple Large Language Models: A Survey on LLM Ensemble","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11301","citing_title":"LatentRouter: Can We Choose the Right Multimodal Model Before Seeing Its Answer?","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09121","citing_title":"A Communication-Theoretic Framework for LLM Agents: Cost-Aware Adaptive Reliability","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06339","citing_title":"A Regime Theory of Controller Class Selection for LLM Action Decisions","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07805","citing_title":"Flexible Routing via Uncertainty Decomposition","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/64LSJYUEBNM4XBDFFQ3JYJT2NB","json":"https://pith.science/pith/64LSJYUEBNM4XBDFFQ3JYJT2NB.json","graph_json":"https://pith.science/api/pith-number/64LSJYUEBNM4XBDFFQ3JYJT2NB/graph.json","events_json":"https://pith.science/api/pith-number/64LSJYUEBNM4XBDFFQ3JYJT2NB/events.json","paper":"https://pith.science/paper/64LSJYUE"},"agent_actions":{"view_html":"https://pith.science/pith/64LSJYUEBNM4XBDFFQ3JYJT2NB","download_json":"https://pith.science/pith/64LSJYUEBNM4XBDFFQ3JYJT2NB.json","view_paper":"https://pith.science/paper/64LSJYUE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.10347&json=true","fetch_graph":"https://pith.science/api/pith-number/64LSJYUEBNM4XBDFFQ3JYJT2NB/graph.json","fetch_events":"https://pith.science/api/pith-number/64LSJYUEBNM4XBDFFQ3JYJT2NB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/64LSJYUEBNM4XBDFFQ3JYJT2NB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/64LSJYUEBNM4XBDFFQ3JYJT2NB/action/storage_attestation","attest_author":"https://pith.science/pith/64LSJYUEBNM4XBDFFQ3JYJT2NB/action/author_attestation","sign_citation":"https://pith.science/pith/64LSJYUEBNM4XBDFFQ3JYJT2NB/action/citation_signature","submit_replication":"https://pith.science/pith/64LSJYUEBNM4XBDFFQ3JYJT2NB/action/replication_record"}},"created_at":"2026-07-05T11:07:15.533865+00:00","updated_at":"2026-07-05T11:07:15.533865+00:00"}