{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:NCYDSJEDP3VZTSOIJV3RCU3TCW","short_pith_number":"pith:NCYDSJED","canonical_record":{"source":{"id":"2601.21349","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-01-29T07:18:33Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"242bf9a45d10ed4f2d12013d1c8f0965812c6006c27be8a40561b453f1875a72","abstract_canon_sha256":"57247740a429ef8809cedf4b48fc4d8a3688f1b6917c1e7097ae9eefffd00194"},"schema_version":"1.0"},"canonical_sha256":"68b03924837eeb99c9c84d7711537315a57b7d7505fac923aad55b81a26547fe","source":{"kind":"arxiv","id":"2601.21349","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2601.21349","created_at":"2026-05-17T23:39:16Z"},{"alias_kind":"arxiv_version","alias_value":"2601.21349v2","created_at":"2026-05-17T23:39:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.21349","created_at":"2026-05-17T23:39:16Z"},{"alias_kind":"pith_short_12","alias_value":"NCYDSJEDP3VZ","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"NCYDSJEDP3VZTSOI","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"NCYDSJED","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:NCYDSJEDP3VZTSOIJV3RCU3TCW","target":"record","payload":{"canonical_record":{"source":{"id":"2601.21349","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-01-29T07:18:33Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"242bf9a45d10ed4f2d12013d1c8f0965812c6006c27be8a40561b453f1875a72","abstract_canon_sha256":"57247740a429ef8809cedf4b48fc4d8a3688f1b6917c1e7097ae9eefffd00194"},"schema_version":"1.0"},"canonical_sha256":"68b03924837eeb99c9c84d7711537315a57b7d7505fac923aad55b81a26547fe","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:16.546085Z","signature_b64":"+Kw2b0UR3gfiuXeB5L8jp8vQZmfrYbRz92Nc9EdmYVLAj2ZPC/ehRfMdh4vXH9+mwjutGEjoXluBpb7W8EH+Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"68b03924837eeb99c9c84d7711537315a57b7d7505fac923aad55b81a26547fe","last_reissued_at":"2026-05-17T23:39:16.545231Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:16.545231Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2601.21349","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:16Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kp70RdIj4PMmo5HRkFkYqFP8Cc7Aarm8neN9J2eMyZgMRfa5mT1SxeiQydw2VbKJJK1oHXxSRN2bcF0s9u9SAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T14:43:29.209180Z"},"content_sha256":"d986f094dab357894cd26fb2bbfc49f537840b1e73e2018445c15d913e4cb614","schema_version":"1.0","event_id":"sha256:d986f094dab357894cd26fb2bbfc49f537840b1e73e2018445c15d913e4cb614"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:NCYDSJEDP3VZTSOIJV3RCU3TCW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"L2R: Low-Rank and Lipschitz-Controlled Routing for Mixture-of-Experts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Projecting mixture-of-experts routing to a low-rank latent space with saturated inner-product scoring yields smoother geometry and stronger expert specialization.","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Guang Li, Miki Haseyama, Minghao Yang, Ren Togo, Takahiro Ogawa","submitted_at":"2026-01-29T07:18:33Z","abstract_excerpt":"Mixture-of-Experts (MoE) models scale neural networks by conditionally activating a small subset of experts, where the router plays a central role in determining expert specialization and overall model performance. However, many modern MoE systems still adopt linear routers in raw high-dimensional representation spaces, where representation mismatch, angular concentration, and scale-sensitive scoring can jointly undermine routing discriminability and stable expert specialization. In this work, we propose Low-rank & Lipschitz-controlled Routing (L2R), a unified routing framework that reshapes b"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"L2R consistently improves routing geometry, expert discrimination, and overall model performance on OLMoE-based language models and ImageNet vision MoE settings.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That projecting to a low-rank latent space plus saturated inner-product scoring preserves sufficient expressiveness while only improving stability, without hidden costs in specialization or generalization.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"L2R improves MoE performance by routing in a low-rank space with Lipschitz-controlled saturated inner-product scoring and multi-anchor mechanisms.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Projecting mixture-of-experts routing to a low-rank latent space with saturated inner-product scoring yields smoother geometry and stronger expert specialization.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"dfbca722c023f096f5c40604e7888af8da156f7471bd404aff2ce5a4649e8226"},"source":{"id":"2601.21349","kind":"arxiv","version":2},"verdict":{"id":"1319a854-684a-495e-bd0a-4cc1ffca5d3a","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T09:53:56.842045Z","strongest_claim":"L2R consistently improves routing geometry, expert discrimination, and overall model performance on OLMoE-based language models and ImageNet vision MoE settings.","one_line_summary":"L2R improves MoE performance by routing in a low-rank space with Lipschitz-controlled saturated inner-product scoring and multi-anchor mechanisms.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That projecting to a low-rank latent space plus saturated inner-product scoring preserves sufficient expressiveness while only improving stability, without hidden costs in specialization or generalization.","pith_extraction_headline":"Projecting mixture-of-experts routing to a low-rank latent space with saturated inner-product scoring yields smoother geometry and stronger expert specialization."},"references":{"count":19,"sample":[{"doi":"","year":1905,"title":"BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions","work_id":"511eeb84-4b95-46d5-b14f-50da43f4f19f","ref_index":1,"cited_arxiv_id":"1905.10044","is_internal_anchor":true},{"doi":"","year":null,"title":"Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge","work_id":"28ea1282-d657-4c61-a83c-f1249be6d6b1","ref_index":2,"cited_arxiv_id":"1803.05457","is_internal_anchor":true},{"doi":"","year":null,"title":"DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models","work_id":"a9888d6d-bf47-4324-9834-7cc12ac3a78c","ref_index":3,"cited_arxiv_id":"2401.06066","is_internal_anchor":true},{"doi":"","year":null,"title":"Toy Models of Superposition","work_id":"43875dbe-bc2d-4ab5-af63-744411533ff7","ref_index":4,"cited_arxiv_id":"2209.10652","is_internal_anchor":true},{"doi":"","year":null,"title":"Mixtral of Experts","work_id":"0de8c352-9daa-4e1e-8c7b-3d0dec69f369","ref_index":5,"cited_arxiv_id":"2401.04088","is_internal_anchor":true}],"resolved_work":19,"snapshot_sha256":"bb7ec696bafdc431a09b60052fffd05a9eaa8473f6cb9dbef9535d979314ecc7","internal_anchors":10},"formal_canon":{"evidence_count":2,"snapshot_sha256":"a30533bc8974317261516c655c679397f50fec86035a63bb5d5fc902dd26b687"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"1319a854-684a-495e-bd0a-4cc1ffca5d3a"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:16Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6S8Smsjb6zEOzBDb4TJGxQZmejf6EzUErHmnijW0YN7SfyegawQCI6L67zC3LkaTTo6StPW0Ea83wRm/hXOgDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T14:43:29.209773Z"},"content_sha256":"3867649dfda971e5226ecba414df741aad70680a1652618744beaefa5c369565","schema_version":"1.0","event_id":"sha256:3867649dfda971e5226ecba414df741aad70680a1652618744beaefa5c369565"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NCYDSJEDP3VZTSOIJV3RCU3TCW/bundle.json","state_url":"https://pith.science/pith/NCYDSJEDP3VZTSOIJV3RCU3TCW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NCYDSJEDP3VZTSOIJV3RCU3TCW/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-27T14:43:29Z","links":{"resolver":"https://pith.science/pith/NCYDSJEDP3VZTSOIJV3RCU3TCW","bundle":"https://pith.science/pith/NCYDSJEDP3VZTSOIJV3RCU3TCW/bundle.json","state":"https://pith.science/pith/NCYDSJEDP3VZTSOIJV3RCU3TCW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NCYDSJEDP3VZTSOIJV3RCU3TCW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:NCYDSJEDP3VZTSOIJV3RCU3TCW","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":"57247740a429ef8809cedf4b48fc4d8a3688f1b6917c1e7097ae9eefffd00194","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-01-29T07:18:33Z","title_canon_sha256":"242bf9a45d10ed4f2d12013d1c8f0965812c6006c27be8a40561b453f1875a72"},"schema_version":"1.0","source":{"id":"2601.21349","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2601.21349","created_at":"2026-05-17T23:39:16Z"},{"alias_kind":"arxiv_version","alias_value":"2601.21349v2","created_at":"2026-05-17T23:39:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.21349","created_at":"2026-05-17T23:39:16Z"},{"alias_kind":"pith_short_12","alias_value":"NCYDSJEDP3VZ","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"NCYDSJEDP3VZTSOI","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"NCYDSJED","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:3867649dfda971e5226ecba414df741aad70680a1652618744beaefa5c369565","target":"graph","created_at":"2026-05-17T23:39:16Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"L2R consistently improves routing geometry, expert discrimination, and overall model performance on OLMoE-based language models and ImageNet vision MoE settings."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That projecting to a low-rank latent space plus saturated inner-product scoring preserves sufficient expressiveness while only improving stability, without hidden costs in specialization or generalization."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"L2R improves MoE performance by routing in a low-rank space with Lipschitz-controlled saturated inner-product scoring and multi-anchor mechanisms."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Projecting mixture-of-experts routing to a low-rank latent space with saturated inner-product scoring yields smoother geometry and stronger expert specialization."}],"snapshot_sha256":"dfbca722c023f096f5c40604e7888af8da156f7471bd404aff2ce5a4649e8226"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"a30533bc8974317261516c655c679397f50fec86035a63bb5d5fc902dd26b687"},"paper":{"abstract_excerpt":"Mixture-of-Experts (MoE) models scale neural networks by conditionally activating a small subset of experts, where the router plays a central role in determining expert specialization and overall model performance. However, many modern MoE systems still adopt linear routers in raw high-dimensional representation spaces, where representation mismatch, angular concentration, and scale-sensitive scoring can jointly undermine routing discriminability and stable expert specialization. In this work, we propose Low-rank & Lipschitz-controlled Routing (L2R), a unified routing framework that reshapes b","authors_text":"Guang Li, Miki Haseyama, Minghao Yang, Ren Togo, Takahiro Ogawa","cross_cats":["cs.AI"],"headline":"Projecting mixture-of-experts routing to a low-rank latent space with saturated inner-product scoring yields smoother geometry and stronger expert specialization.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-01-29T07:18:33Z","title":"L2R: Low-Rank and Lipschitz-Controlled Routing for Mixture-of-Experts"},"references":{"count":19,"internal_anchors":10,"resolved_work":19,"sample":[{"cited_arxiv_id":"1905.10044","doi":"","is_internal_anchor":true,"ref_index":1,"title":"BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions","work_id":"511eeb84-4b95-46d5-b14f-50da43f4f19f","year":1905},{"cited_arxiv_id":"1803.05457","doi":"","is_internal_anchor":true,"ref_index":2,"title":"Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge","work_id":"28ea1282-d657-4c61-a83c-f1249be6d6b1","year":null},{"cited_arxiv_id":"2401.06066","doi":"","is_internal_anchor":true,"ref_index":3,"title":"DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models","work_id":"a9888d6d-bf47-4324-9834-7cc12ac3a78c","year":null},{"cited_arxiv_id":"2209.10652","doi":"","is_internal_anchor":true,"ref_index":4,"title":"Toy Models of Superposition","work_id":"43875dbe-bc2d-4ab5-af63-744411533ff7","year":null},{"cited_arxiv_id":"2401.04088","doi":"","is_internal_anchor":true,"ref_index":5,"title":"Mixtral of Experts","work_id":"0de8c352-9daa-4e1e-8c7b-3d0dec69f369","year":null}],"snapshot_sha256":"bb7ec696bafdc431a09b60052fffd05a9eaa8473f6cb9dbef9535d979314ecc7"},"source":{"id":"2601.21349","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-16T09:53:56.842045Z","id":"1319a854-684a-495e-bd0a-4cc1ffca5d3a","model_set":{"reader":"grok-4.3"},"one_line_summary":"L2R improves MoE performance by routing in a low-rank space with Lipschitz-controlled saturated inner-product scoring and multi-anchor mechanisms.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Projecting mixture-of-experts routing to a low-rank latent space with saturated inner-product scoring yields smoother geometry and stronger expert specialization.","strongest_claim":"L2R consistently improves routing geometry, expert discrimination, and overall model performance on OLMoE-based language models and ImageNet vision MoE settings.","weakest_assumption":"That projecting to a low-rank latent space plus saturated inner-product scoring preserves sufficient expressiveness while only improving stability, without hidden costs in specialization or generalization."}},"verdict_id":"1319a854-684a-495e-bd0a-4cc1ffca5d3a"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:d986f094dab357894cd26fb2bbfc49f537840b1e73e2018445c15d913e4cb614","target":"record","created_at":"2026-05-17T23:39:16Z","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":"57247740a429ef8809cedf4b48fc4d8a3688f1b6917c1e7097ae9eefffd00194","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-01-29T07:18:33Z","title_canon_sha256":"242bf9a45d10ed4f2d12013d1c8f0965812c6006c27be8a40561b453f1875a72"},"schema_version":"1.0","source":{"id":"2601.21349","kind":"arxiv","version":2}},"canonical_sha256":"68b03924837eeb99c9c84d7711537315a57b7d7505fac923aad55b81a26547fe","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"68b03924837eeb99c9c84d7711537315a57b7d7505fac923aad55b81a26547fe","first_computed_at":"2026-05-17T23:39:16.545231Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:16.545231Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"+Kw2b0UR3gfiuXeB5L8jp8vQZmfrYbRz92Nc9EdmYVLAj2ZPC/ehRfMdh4vXH9+mwjutGEjoXluBpb7W8EH+Bg==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:16.546085Z","signed_message":"canonical_sha256_bytes"},"source_id":"2601.21349","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d986f094dab357894cd26fb2bbfc49f537840b1e73e2018445c15d913e4cb614","sha256:3867649dfda971e5226ecba414df741aad70680a1652618744beaefa5c369565"],"state_sha256":"b0c672d25d5c604dbf956c8deb15bb97ee1f83c886079ccab3988553890503f8"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jZc2z0yUcUR36RyHJH+l5yEsgu65OS8oYk1BQMCUaxWxaa9c4D3wfLYuJgZNJnrf3aEfUcP+QheC2fkiq2B4CQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T14:43:29.213212Z","bundle_sha256":"81d81aeedda528d31b7958c20c1d2ff4ea720e175f703642ea8026fc7614552a"}}