{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:2RYLWIK5THFDECVGWLE7QT4V7R","short_pith_number":"pith:2RYLWIK5","schema_version":"1.0","canonical_sha256":"d470bb215d99ca320aa6b2c9f84f95fc552b2e8ce85fa587053d1ce683c0da4f","source":{"kind":"arxiv","id":"2605.30876","version":1},"attestation_state":"computed","paper":{"title":"dMoE: dLLMs with Learnable Block Experts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Gongfan Fang, Sicheng Feng, Xinchao Wang, Xinyin Ma, Zigeng Chen","submitted_at":"2026-05-29T06:03:50Z","abstract_excerpt":"Diffusion Large Language Models (dLLMs) have recently emerged as a promising alternative to autoregressive models, offering competitive performance while naturally supporting parallel decoding. However, as dLLMs are increasingly integrated with Mixture-of-Experts (MoE) architectures to scale model capacity, a fundamental mismatch arises between block parallel decoding and token-level expert selection. Specifically, each dLLM forward pass processes multiple tokens with bidirectional dependencies, whereas conventional MoE layers route each token independently. This mismatch substantially increas"},"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":"2605.30876","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-29T06:03:50Z","cross_cats_sorted":[],"title_canon_sha256":"6367c86f36beb5ebd08fd1313e5212212b124ff3bde793996c8ba6cf98fd083a","abstract_canon_sha256":"7d29557179ed3bc1344967c4fdb83b5d353940a0c8eaf8fc55b07249f05b2288"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-01T01:03:22.466388Z","signature_b64":"4LdNx3ltfeemlr0Ioh/N3xPNYDF9jM0MNCGihzYvc8upGA1e/ew4FMkgPop9IppeJ8AHdm3N3bZc1J9GGJCvCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d470bb215d99ca320aa6b2c9f84f95fc552b2e8ce85fa587053d1ce683c0da4f","last_reissued_at":"2026-06-01T01:03:22.465722Z","signature_status":"signed_v1","first_computed_at":"2026-06-01T01:03:22.465722Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"dMoE: dLLMs with Learnable Block Experts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Gongfan Fang, Sicheng Feng, Xinchao Wang, Xinyin Ma, Zigeng Chen","submitted_at":"2026-05-29T06:03:50Z","abstract_excerpt":"Diffusion Large Language Models (dLLMs) have recently emerged as a promising alternative to autoregressive models, offering competitive performance while naturally supporting parallel decoding. However, as dLLMs are increasingly integrated with Mixture-of-Experts (MoE) architectures to scale model capacity, a fundamental mismatch arises between block parallel decoding and token-level expert selection. Specifically, each dLLM forward pass processes multiple tokens with bidirectional dependencies, whereas conventional MoE layers route each token independently. This mismatch substantially increas"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.30876","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/2605.30876/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":"2605.30876","created_at":"2026-06-01T01:03:22.465845+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.30876v1","created_at":"2026-06-01T01:03:22.465845+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.30876","created_at":"2026-06-01T01:03:22.465845+00:00"},{"alias_kind":"pith_short_12","alias_value":"2RYLWIK5THFD","created_at":"2026-06-01T01:03:22.465845+00:00"},{"alias_kind":"pith_short_16","alias_value":"2RYLWIK5THFDECVG","created_at":"2026-06-01T01:03:22.465845+00:00"},{"alias_kind":"pith_short_8","alias_value":"2RYLWIK5","created_at":"2026-06-01T01:03:22.465845+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/2RYLWIK5THFDECVGWLE7QT4V7R","json":"https://pith.science/pith/2RYLWIK5THFDECVGWLE7QT4V7R.json","graph_json":"https://pith.science/api/pith-number/2RYLWIK5THFDECVGWLE7QT4V7R/graph.json","events_json":"https://pith.science/api/pith-number/2RYLWIK5THFDECVGWLE7QT4V7R/events.json","paper":"https://pith.science/paper/2RYLWIK5"},"agent_actions":{"view_html":"https://pith.science/pith/2RYLWIK5THFDECVGWLE7QT4V7R","download_json":"https://pith.science/pith/2RYLWIK5THFDECVGWLE7QT4V7R.json","view_paper":"https://pith.science/paper/2RYLWIK5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.30876&json=true","fetch_graph":"https://pith.science/api/pith-number/2RYLWIK5THFDECVGWLE7QT4V7R/graph.json","fetch_events":"https://pith.science/api/pith-number/2RYLWIK5THFDECVGWLE7QT4V7R/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2RYLWIK5THFDECVGWLE7QT4V7R/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2RYLWIK5THFDECVGWLE7QT4V7R/action/storage_attestation","attest_author":"https://pith.science/pith/2RYLWIK5THFDECVGWLE7QT4V7R/action/author_attestation","sign_citation":"https://pith.science/pith/2RYLWIK5THFDECVGWLE7QT4V7R/action/citation_signature","submit_replication":"https://pith.science/pith/2RYLWIK5THFDECVGWLE7QT4V7R/action/replication_record"}},"created_at":"2026-06-01T01:03:22.465845+00:00","updated_at":"2026-06-01T01:03:22.465845+00:00"}