{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:CSFO6H4736JODECKN3JP2ZJSWM","short_pith_number":"pith:CSFO6H47","schema_version":"1.0","canonical_sha256":"148aef1f9fdf92e1904a6ed2fd6532b330e0a1f9fdeab4517cdec6ec7e33a3ce","source":{"kind":"arxiv","id":"2601.02144","version":2},"attestation_state":"computed","paper":{"title":"Routing by Analogy: kNN-Augmented Expert Assignment for Mixture-of-Experts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Boxuan Lyu, Hidetaka Kamigaito, Peinan Zhang, Soichiro Murakami","submitted_at":"2026-01-05T14:16:11Z","abstract_excerpt":"Mixture-of-Experts (MoE) architectures scale large language models efficiently by employing a parametric ``router'' to dispatch tokens to a sparse subset of experts. Typically, this router is trained once and then frozen, rendering routing decisions brittle under distribution shifts. We address this limitation by introducing kNN-MoE, a retrieval-augmented routing framework that reuses locally optimal expert assignments from a memory of similar past cases. This memory is constructed offline by directly optimizing token-wise routing logits to maximize the likelihood on a reference set. Crucially"},"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":"2601.02144","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-01-05T14:16:11Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"e7f401d9d20255dcc2687424e0c415a0f40ffde00ba14afb02dd41758bc65ffd","abstract_canon_sha256":"efddf5494f36d7a5fdb996ccba3a05a18d7865591c95c146cb8c4255fa921928"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T02:04:02.298797Z","signature_b64":"9uVeKF/nIFmDLAKZBQPmarc5Zd8UvxQSii7w135crfD/4Ct2mmttdkElplAyAsUGDDDB2i9b74zCXF4+UgxiBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"148aef1f9fdf92e1904a6ed2fd6532b330e0a1f9fdeab4517cdec6ec7e33a3ce","last_reissued_at":"2026-05-26T02:04:02.297857Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T02:04:02.297857Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Routing by Analogy: kNN-Augmented Expert Assignment for Mixture-of-Experts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Boxuan Lyu, Hidetaka Kamigaito, Peinan Zhang, Soichiro Murakami","submitted_at":"2026-01-05T14:16:11Z","abstract_excerpt":"Mixture-of-Experts (MoE) architectures scale large language models efficiently by employing a parametric ``router'' to dispatch tokens to a sparse subset of experts. Typically, this router is trained once and then frozen, rendering routing decisions brittle under distribution shifts. We address this limitation by introducing kNN-MoE, a retrieval-augmented routing framework that reuses locally optimal expert assignments from a memory of similar past cases. This memory is constructed offline by directly optimizing token-wise routing logits to maximize the likelihood on a reference set. Crucially"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.02144","kind":"arxiv","version":2},"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/2601.02144/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":"2601.02144","created_at":"2026-05-26T02:04:02.297997+00:00"},{"alias_kind":"arxiv_version","alias_value":"2601.02144v2","created_at":"2026-05-26T02:04:02.297997+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.02144","created_at":"2026-05-26T02:04:02.297997+00:00"},{"alias_kind":"pith_short_12","alias_value":"CSFO6H4736JO","created_at":"2026-05-26T02:04:02.297997+00:00"},{"alias_kind":"pith_short_16","alias_value":"CSFO6H4736JODECK","created_at":"2026-05-26T02:04:02.297997+00:00"},{"alias_kind":"pith_short_8","alias_value":"CSFO6H47","created_at":"2026-05-26T02:04:02.297997+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/CSFO6H4736JODECKN3JP2ZJSWM","json":"https://pith.science/pith/CSFO6H4736JODECKN3JP2ZJSWM.json","graph_json":"https://pith.science/api/pith-number/CSFO6H4736JODECKN3JP2ZJSWM/graph.json","events_json":"https://pith.science/api/pith-number/CSFO6H4736JODECKN3JP2ZJSWM/events.json","paper":"https://pith.science/paper/CSFO6H47"},"agent_actions":{"view_html":"https://pith.science/pith/CSFO6H4736JODECKN3JP2ZJSWM","download_json":"https://pith.science/pith/CSFO6H4736JODECKN3JP2ZJSWM.json","view_paper":"https://pith.science/paper/CSFO6H47","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2601.02144&json=true","fetch_graph":"https://pith.science/api/pith-number/CSFO6H4736JODECKN3JP2ZJSWM/graph.json","fetch_events":"https://pith.science/api/pith-number/CSFO6H4736JODECKN3JP2ZJSWM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CSFO6H4736JODECKN3JP2ZJSWM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CSFO6H4736JODECKN3JP2ZJSWM/action/storage_attestation","attest_author":"https://pith.science/pith/CSFO6H4736JODECKN3JP2ZJSWM/action/author_attestation","sign_citation":"https://pith.science/pith/CSFO6H4736JODECKN3JP2ZJSWM/action/citation_signature","submit_replication":"https://pith.science/pith/CSFO6H4736JODECKN3JP2ZJSWM/action/replication_record"}},"created_at":"2026-05-26T02:04:02.297997+00:00","updated_at":"2026-05-26T02:04:02.297997+00:00"}