{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:4JIKGC4NWX2TSZM7QRNYUWUQUV","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"6243323ea3b1db6c1bf88dede7488239d97dd8de96d9998a17178e87c654c939","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-12T20:30:28Z","title_canon_sha256":"b485ff06966b09029838f70a41ea2b53809ec3ef0e8c125f95cc40c8d8a38648"},"schema_version":"1.0","source":{"id":"2502.08773","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.08773","created_at":"2026-07-05T11:41:21Z"},{"alias_kind":"arxiv_version","alias_value":"2502.08773v2","created_at":"2026-07-05T11:41:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.08773","created_at":"2026-07-05T11:41:21Z"},{"alias_kind":"pith_short_12","alias_value":"4JIKGC4NWX2T","created_at":"2026-07-05T11:41:21Z"},{"alias_kind":"pith_short_16","alias_value":"4JIKGC4NWX2TSZM7","created_at":"2026-07-05T11:41:21Z"},{"alias_kind":"pith_short_8","alias_value":"4JIKGC4N","created_at":"2026-07-05T11:41:21Z"}],"graph_snapshots":[{"event_id":"sha256:67b05576fc25d3c89bf9d227eb002257c7600ceadca1f2a3c9aa9d77cf5b10d6","target":"graph","created_at":"2026-07-05T11:41:21Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2502.08773/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Model routing is a simple technique for reducing the inference cost of large language models (LLMs), wherein one maintains a pool of candidate LLMs, and learns to route each prompt to the smallest feasible LLM. Existing works focus on learning a router for a fixed pool of LLMs. In this paper, we consider the problem of dynamic routing, where new, previously unobserved LLMs are available at test time. We propose UniRoute, a new approach to this problem that relies on representing each LLM as a feature vector, derived based on predictions on a set of representative prompts. Based on this, we det","authors_text":"Aditya Krishna Menon, Alec Go, Ankit Singh Rawat, Chen-Yu Lee, Congchao Wang, Harikrishna Narasimhan, Jeevesh Juneja, Pradeep Shenoy, Rina Panigrahy, Sanjiv Kumar, Wittawat Jitkrittum, Zifeng Wang","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-12T20:30:28Z","title":"Universal Model Routing for Efficient LLM Inference"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.08773","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:f140516668e3857a53af85da8e43ef0a922d1170660381f82557925afac523d8","target":"record","created_at":"2026-07-05T11:41:21Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"6243323ea3b1db6c1bf88dede7488239d97dd8de96d9998a17178e87c654c939","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-12T20:30:28Z","title_canon_sha256":"b485ff06966b09029838f70a41ea2b53809ec3ef0e8c125f95cc40c8d8a38648"},"schema_version":"1.0","source":{"id":"2502.08773","kind":"arxiv","version":2}},"canonical_sha256":"e250a30b8db5f539659f845b8a5a90a575a148cb59345e42130f317fbb2c2198","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e250a30b8db5f539659f845b8a5a90a575a148cb59345e42130f317fbb2c2198","first_computed_at":"2026-07-05T11:41:21.393976Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:41:21.393976Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"luyrCNAOnNhWIdHrO0rXioEEo7HQ/AHnyiwUQvbEE1bgEQI/VMk/qyo4EHo7u6Ym9FK9adUosfjg/OhlropVCA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:41:21.394523Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.08773","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f140516668e3857a53af85da8e43ef0a922d1170660381f82557925afac523d8","sha256:67b05576fc25d3c89bf9d227eb002257c7600ceadca1f2a3c9aa9d77cf5b10d6"],"state_sha256":"b85c254116a1c885ce2d9d9ede0b3ce361144e359982fcdbf294a0c43cf4c8ac"}