{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:M4DLISHPK4V3ZY5KWLBILQCT4M","short_pith_number":"pith:M4DLISHP","schema_version":"1.0","canonical_sha256":"6706b448ef572bbce3aab2c285c053e3170329673d71ea06e9a43b24ada2af0e","source":{"kind":"arxiv","id":"2402.01739","version":2},"attestation_state":"computed","paper":{"title":"OpenMoE: An Early Effort on Open Mixture-of-Experts Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.DC","cs.LG"],"primary_cat":"cs.CL","authors_text":"Fuzhao Xue, Jinjie Ni, Wangchunshu Zhou, Yang You, Yao Fu, Zangwei Zheng, Zian Zheng","submitted_at":"2024-01-29T12:05:02Z","abstract_excerpt":"To help the open-source community have a better understanding of Mixture-of-Experts (MoE) based large language models (LLMs), we train and release OpenMoE, a series of fully open-sourced and reproducible decoder-only MoE LLMs, ranging from 650M to 34B parameters and trained on up to over 1T tokens. Our investigation confirms that MoE-based LLMs can offer a more favorable cost-effectiveness trade-off than dense LLMs, highlighting the potential effectiveness for future LLM development.\n  One more important contribution of this study is an in-depth analysis of the routing mechanisms within our Op"},"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":"2402.01739","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-29T12:05:02Z","cross_cats_sorted":["cs.AI","cs.DC","cs.LG"],"title_canon_sha256":"977d9cb232534d31c9cb18bf880e2f8b3b4014585f727ab960d859cc4b5dea01","abstract_canon_sha256":"3c2558a16b6e4fa46bb4abcfe07ec67d117fb221ab861e69e72fec08813e5d4b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:01:18.613003Z","signature_b64":"uNsQWsThnWQl9Rj0wB985lfVzh+dbWXtLXwtH+wcOZTRxqAaDcf22Jlo6bmhJHZg8Jr9FBXO1CK0LqHjN4ubAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6706b448ef572bbce3aab2c285c053e3170329673d71ea06e9a43b24ada2af0e","last_reissued_at":"2026-07-05T08:01:18.612419Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:01:18.612419Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"OpenMoE: An Early Effort on Open Mixture-of-Experts Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.DC","cs.LG"],"primary_cat":"cs.CL","authors_text":"Fuzhao Xue, Jinjie Ni, Wangchunshu Zhou, Yang You, Yao Fu, Zangwei Zheng, Zian Zheng","submitted_at":"2024-01-29T12:05:02Z","abstract_excerpt":"To help the open-source community have a better understanding of Mixture-of-Experts (MoE) based large language models (LLMs), we train and release OpenMoE, a series of fully open-sourced and reproducible decoder-only MoE LLMs, ranging from 650M to 34B parameters and trained on up to over 1T tokens. Our investigation confirms that MoE-based LLMs can offer a more favorable cost-effectiveness trade-off than dense LLMs, highlighting the potential effectiveness for future LLM development.\n  One more important contribution of this study is an in-depth analysis of the routing mechanisms within our Op"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.01739","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/2402.01739/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":"2402.01739","created_at":"2026-07-05T08:01:18.612490+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.01739v2","created_at":"2026-07-05T08:01:18.612490+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.01739","created_at":"2026-07-05T08:01:18.612490+00:00"},{"alias_kind":"pith_short_12","alias_value":"M4DLISHPK4V3","created_at":"2026-07-05T08:01:18.612490+00:00"},{"alias_kind":"pith_short_16","alias_value":"M4DLISHPK4V3ZY5K","created_at":"2026-07-05T08:01:18.612490+00:00"},{"alias_kind":"pith_short_8","alias_value":"M4DLISHP","created_at":"2026-07-05T08:01:18.612490+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":23,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.08077","citing_title":"Modular Pretraining Enables Access Control","ref_index":82,"is_internal_anchor":true},{"citing_arxiv_id":"2606.26287","citing_title":"GeMoE: Gating Entropy is All You Need for Uncertainty-aware Adaptive Routing in MoE-based Large Vision-Language Models","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2606.21645","citing_title":"Behavioral and Representational Evidence of Binomial Ordering Preferences in Large Language Models","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2606.21228","citing_title":"Sakana Fugu Technical Report","ref_index":222,"is_internal_anchor":false},{"citing_arxiv_id":"2606.19195","citing_title":"Moebius: 0.2B Lightweight Image Inpainting Framework with 10B-Level Performance","ref_index":57,"is_internal_anchor":false},{"citing_arxiv_id":"2606.04438","citing_title":"LoopMoE: Unifying Iterative Computation with Mixture-of-Experts for Language Modeling","ref_index":53,"is_internal_anchor":false},{"citing_arxiv_id":"2606.04160","citing_title":"Expert-Aware Refusal Steering","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01047","citing_title":"Learning Multi-Modal Trajectory Policies for Data-Efficient Robotic Manipulation","ref_index":53,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00275","citing_title":"Hyperbolic and Evidence-Prioritized Experts for Large Vision-Language Models","ref_index":46,"is_internal_anchor":false},{"citing_arxiv_id":"2605.26776","citing_title":"Towards Generalization-Oriented Models for Vehicle Routing Problems with Mixture-of-Experts","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2503.05066","citing_title":"Capacity-Aware Inference: Mitigating the Straggler Effect in Mixture of Experts","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2601.05106","citing_title":"Token-Level LLM Collaboration via FusionRoute","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20610","citing_title":"Beyond Routing: Characterising Expert Tuning and Representation in Vision Mixture-of-Experts","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15484","citing_title":"When Does Sparse MoE Help in Vision? The Role of Backbone Compute Leverage in Sparse Routing","ref_index":57,"is_internal_anchor":false},{"citing_arxiv_id":"2509.19349","citing_title":"ShinkaEvolve: Towards Open-Ended And Sample-Efficient Program Evolution","ref_index":261,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06211","citing_title":"Illocutionary Explanation Planning for Source-Faithful Explanations in Retrieval-Augmented Language Models","ref_index":52,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08292","citing_title":"Hierarchical Mixture-of-Experts with Two-Stage Optimization","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06665","citing_title":"UniPool: A Globally Shared Expert Pool for Mixture-of-Experts","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04449","citing_title":"GEM: Graph-Enhanced Mixture-of-Experts with ReAct Agents for Dialogue State Tracking","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08541","citing_title":"Seeing but Not Thinking: Routing Distraction in Multimodal Mixture-of-Experts","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07260","citing_title":"When Are Experts Misrouted? Counterfactual Routing Analysis in Mixture-of-Experts Language Models","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05225","citing_title":"MACS: Modality-Aware Capacity Scaling for Efficient Multimodal MoE Inference","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05225","citing_title":"MACS: Modality-Aware Capacity Scaling for Efficient Multimodal MoE Inference","ref_index":8,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/M4DLISHPK4V3ZY5KWLBILQCT4M","json":"https://pith.science/pith/M4DLISHPK4V3ZY5KWLBILQCT4M.json","graph_json":"https://pith.science/api/pith-number/M4DLISHPK4V3ZY5KWLBILQCT4M/graph.json","events_json":"https://pith.science/api/pith-number/M4DLISHPK4V3ZY5KWLBILQCT4M/events.json","paper":"https://pith.science/paper/M4DLISHP"},"agent_actions":{"view_html":"https://pith.science/pith/M4DLISHPK4V3ZY5KWLBILQCT4M","download_json":"https://pith.science/pith/M4DLISHPK4V3ZY5KWLBILQCT4M.json","view_paper":"https://pith.science/paper/M4DLISHP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.01739&json=true","fetch_graph":"https://pith.science/api/pith-number/M4DLISHPK4V3ZY5KWLBILQCT4M/graph.json","fetch_events":"https://pith.science/api/pith-number/M4DLISHPK4V3ZY5KWLBILQCT4M/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/M4DLISHPK4V3ZY5KWLBILQCT4M/action/timestamp_anchor","attest_storage":"https://pith.science/pith/M4DLISHPK4V3ZY5KWLBILQCT4M/action/storage_attestation","attest_author":"https://pith.science/pith/M4DLISHPK4V3ZY5KWLBILQCT4M/action/author_attestation","sign_citation":"https://pith.science/pith/M4DLISHPK4V3ZY5KWLBILQCT4M/action/citation_signature","submit_replication":"https://pith.science/pith/M4DLISHPK4V3ZY5KWLBILQCT4M/action/replication_record"}},"created_at":"2026-07-05T08:01:18.612490+00:00","updated_at":"2026-07-05T08:01:18.612490+00:00"}