{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:7OKQMHXYJJZHMJVM6ADTPBOGFY","short_pith_number":"pith:7OKQMHXY","schema_version":"1.0","canonical_sha256":"fb95061ef84a727626acf0073785c62e38f99b9e60a1bd8609e338271a62231f","source":{"kind":"arxiv","id":"2506.18945","version":1},"attestation_state":"computed","paper":{"title":"Chain-of-Experts: Unlocking the Communication Power of Mixture-of-Experts Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Jiajun Wu, Jiarui Yao, Linjie Li, Lu Yin, Manling Li, Robert Csordas, Rui Pan, Shiwei Liu, Tong Zhang, Zihan Wang","submitted_at":"2025-06-23T02:15:43Z","abstract_excerpt":"We propose Chain-of-Experts (CoE), a new Mixture-of-Experts (MoE) architecture that introduces sequential expert communication within each layer. Unlike traditional MoE models, where experts operate independently in parallel, CoE processes tokens iteratively across a chain of experts inside a layer. To support dynamic expert selection across iterations, CoE employs a dedicated router at each iteration step within a layer. This design allows tokens to re-evaluate and select different experts during each iteration, rather than being statically assigned. As a result, CoE introduces a flexible rou"},"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":"2506.18945","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-23T02:15:43Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"b0ac62db2f3cdae92bbbb051f8f6b633f1aa7dac052323da3fe4656d08c61021","abstract_canon_sha256":"e391efbdb3d09175a901b9c315612ede2da8cf3b82cfde811283617c874c8b27"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:26:06.981616Z","signature_b64":"HAru4gS79drDVV06Qvr+qi/tKpAfFeWJZJpeQTOPmiV3CVASvvJvDkZiuSB7rKP//stWuCS0RaX7QtZ54stZCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fb95061ef84a727626acf0073785c62e38f99b9e60a1bd8609e338271a62231f","last_reissued_at":"2026-07-05T11:26:06.981101Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:26:06.981101Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Chain-of-Experts: Unlocking the Communication Power of Mixture-of-Experts Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Jiajun Wu, Jiarui Yao, Linjie Li, Lu Yin, Manling Li, Robert Csordas, Rui Pan, Shiwei Liu, Tong Zhang, Zihan Wang","submitted_at":"2025-06-23T02:15:43Z","abstract_excerpt":"We propose Chain-of-Experts (CoE), a new Mixture-of-Experts (MoE) architecture that introduces sequential expert communication within each layer. Unlike traditional MoE models, where experts operate independently in parallel, CoE processes tokens iteratively across a chain of experts inside a layer. To support dynamic expert selection across iterations, CoE employs a dedicated router at each iteration step within a layer. This design allows tokens to re-evaluate and select different experts during each iteration, rather than being statically assigned. As a result, CoE introduces a flexible rou"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.18945","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/2506.18945/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":"2506.18945","created_at":"2026-07-05T11:26:06.981165+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.18945v1","created_at":"2026-07-05T11:26:06.981165+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.18945","created_at":"2026-07-05T11:26:06.981165+00:00"},{"alias_kind":"pith_short_12","alias_value":"7OKQMHXYJJZH","created_at":"2026-07-05T11:26:06.981165+00:00"},{"alias_kind":"pith_short_16","alias_value":"7OKQMHXYJJZHMJVM","created_at":"2026-07-05T11:26:06.981165+00:00"},{"alias_kind":"pith_short_8","alias_value":"7OKQMHXY","created_at":"2026-07-05T11:26:06.981165+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.18089","citing_title":"From Reasoning Traces to Reusable Modules: Understanding Compositional Generalization in Language Model Reasoning","ref_index":97,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7OKQMHXYJJZHMJVM6ADTPBOGFY","json":"https://pith.science/pith/7OKQMHXYJJZHMJVM6ADTPBOGFY.json","graph_json":"https://pith.science/api/pith-number/7OKQMHXYJJZHMJVM6ADTPBOGFY/graph.json","events_json":"https://pith.science/api/pith-number/7OKQMHXYJJZHMJVM6ADTPBOGFY/events.json","paper":"https://pith.science/paper/7OKQMHXY"},"agent_actions":{"view_html":"https://pith.science/pith/7OKQMHXYJJZHMJVM6ADTPBOGFY","download_json":"https://pith.science/pith/7OKQMHXYJJZHMJVM6ADTPBOGFY.json","view_paper":"https://pith.science/paper/7OKQMHXY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.18945&json=true","fetch_graph":"https://pith.science/api/pith-number/7OKQMHXYJJZHMJVM6ADTPBOGFY/graph.json","fetch_events":"https://pith.science/api/pith-number/7OKQMHXYJJZHMJVM6ADTPBOGFY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7OKQMHXYJJZHMJVM6ADTPBOGFY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7OKQMHXYJJZHMJVM6ADTPBOGFY/action/storage_attestation","attest_author":"https://pith.science/pith/7OKQMHXYJJZHMJVM6ADTPBOGFY/action/author_attestation","sign_citation":"https://pith.science/pith/7OKQMHXYJJZHMJVM6ADTPBOGFY/action/citation_signature","submit_replication":"https://pith.science/pith/7OKQMHXYJJZHMJVM6ADTPBOGFY/action/replication_record"}},"created_at":"2026-07-05T11:26:06.981165+00:00","updated_at":"2026-07-05T11:26:06.981165+00:00"}