{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:OCUAVJ6C2NN4A57DNLXOAHBLGJ","short_pith_number":"pith:OCUAVJ6C","schema_version":"1.0","canonical_sha256":"70a80aa7c2d35bc077e36aeee01c2b325eb4a3c5c21093546a8a258f964b1d02","source":{"kind":"arxiv","id":"2505.16278","version":2},"attestation_state":"computed","paper":{"title":"DriveMoE: Mixture-of-Experts for Vision-Language-Action Model in End-to-End Autonomous Driving","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.RO"],"primary_cat":"cs.CV","authors_text":"Haisheng Su, Junchi Yan, Qifeng Li, Xiaosong Jia, Xuekai Zhu, Yilin Chai, Yuqian Shao, Zhenjie Yang","submitted_at":"2025-05-22T06:23:04Z","abstract_excerpt":"End-to-end autonomous driving (E2E-AD) demands effective processing of multi-view sensory data and robust handling of diverse and complex driving scenarios, particularly rare maneuvers such as aggressive turns. Recent success of Mixture-of-Experts (MoE) architecture in Large Language Models (LLMs) demonstrates that specialization of parameters enables strong scalability. In this work, we propose DriveMoE, a novel MoE-based E2E-AD framework, with a Scene-Specialized Vision MoE and a Skill-Specialized Action MoE. DriveMoE is built upon our $\\pi_0$ Vision-Language-Action (VLA) baseline (originall"},"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":"2505.16278","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-22T06:23:04Z","cross_cats_sorted":["cs.AI","cs.RO"],"title_canon_sha256":"bcff7baf6ae1d825fb68d376ef51e21dc652ef395acacaf81fa8e4e2c00be5ec","abstract_canon_sha256":"f542826bb56c23d84733dfb7f5d0ea3bbc5ae9bb448241bd96cece3ceaeec0c4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:05:28.996685Z","signature_b64":"iiZSvZe/KiGZFusiEEVnmCBaPe+Ei0U/SEWEUPCDUZl1wzCzMh8y2Mokot4/yr2mhwcAk9xeau4v/mG5ONcEBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"70a80aa7c2d35bc077e36aeee01c2b325eb4a3c5c21093546a8a258f964b1d02","last_reissued_at":"2026-05-20T00:05:28.995790Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:05:28.995790Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DriveMoE: Mixture-of-Experts for Vision-Language-Action Model in End-to-End Autonomous Driving","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.RO"],"primary_cat":"cs.CV","authors_text":"Haisheng Su, Junchi Yan, Qifeng Li, Xiaosong Jia, Xuekai Zhu, Yilin Chai, Yuqian Shao, Zhenjie Yang","submitted_at":"2025-05-22T06:23:04Z","abstract_excerpt":"End-to-end autonomous driving (E2E-AD) demands effective processing of multi-view sensory data and robust handling of diverse and complex driving scenarios, particularly rare maneuvers such as aggressive turns. Recent success of Mixture-of-Experts (MoE) architecture in Large Language Models (LLMs) demonstrates that specialization of parameters enables strong scalability. In this work, we propose DriveMoE, a novel MoE-based E2E-AD framework, with a Scene-Specialized Vision MoE and a Skill-Specialized Action MoE. DriveMoE is built upon our $\\pi_0$ Vision-Language-Action (VLA) baseline (originall"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.16278","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/2505.16278/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":"2505.16278","created_at":"2026-05-20T00:05:28.995922+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.16278v2","created_at":"2026-05-20T00:05:28.995922+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.16278","created_at":"2026-05-20T00:05:28.995922+00:00"},{"alias_kind":"pith_short_12","alias_value":"OCUAVJ6C2NN4","created_at":"2026-05-20T00:05:28.995922+00:00"},{"alias_kind":"pith_short_16","alias_value":"OCUAVJ6C2NN4A57D","created_at":"2026-05-20T00:05:28.995922+00:00"},{"alias_kind":"pith_short_8","alias_value":"OCUAVJ6C","created_at":"2026-05-20T00:05:28.995922+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":23,"internal_anchor_count":23,"sample":[{"citing_arxiv_id":"2605.23270","citing_title":"ChainFlow-VLA: Causal Flow Planning with Vision-Language Models","ref_index":19,"is_internal_anchor":true},{"citing_arxiv_id":"2512.10719","citing_title":"SpaceDrive: Infusing Spatial Awareness into VLM-based Autonomous Driving","ref_index":69,"is_internal_anchor":true},{"citing_arxiv_id":"2604.24514","citing_title":"SceneSelect: Selective Learning for Trajectory Scene Classification and Expert Scheduling","ref_index":37,"is_internal_anchor":true},{"citing_arxiv_id":"2605.22089","citing_title":"LVDrive: Latent Visual Representation Enhanced Vision-Language-Action Autonomous Driving Model","ref_index":56,"is_internal_anchor":true},{"citing_arxiv_id":"2605.22504","citing_title":"LACO: Adaptive Latent Communication for Collaborative Driving","ref_index":40,"is_internal_anchor":true},{"citing_arxiv_id":"2605.08830","citing_title":"VECTOR-Drive: Tightly Coupled Vision-Language and Trajectory Expert Routing for End-to-End Autonomous Driving","ref_index":26,"is_internal_anchor":true},{"citing_arxiv_id":"2605.17907","citing_title":"One Model to Translate Them All: Universal Any-to-Any Translation for Heterogeneous Collaborative Perception","ref_index":42,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18059","citing_title":"Bench2Drive-Robust: Benchmarking Closed-Loop Autonomous Driving under Deployment Perturbations","ref_index":50,"is_internal_anchor":true},{"citing_arxiv_id":"2506.09981","citing_title":"ReSim: Reliable World Simulation for Autonomous Driving","ref_index":97,"is_internal_anchor":true},{"citing_arxiv_id":"2510.12796","citing_title":"DriveVLA-W0: World Models Amplify Data Scaling Law in Autonomous Driving","ref_index":30,"is_internal_anchor":true},{"citing_arxiv_id":"2511.18085","citing_title":"Continually Evolving Skill Knowledge in Vision Language Action Model","ref_index":46,"is_internal_anchor":true},{"citing_arxiv_id":"2601.07060","citing_title":"PALM: Progress-Aware Policy Learning via Affordance Reasoning for Long-Horizon Robotic Manipulation","ref_index":128,"is_internal_anchor":true},{"citing_arxiv_id":"2603.18561","citing_title":"CausalVAD: De-confounding End-to-End Autonomous Driving via Causal Intervention","ref_index":38,"is_internal_anchor":true},{"citing_arxiv_id":"2605.14696","citing_title":"EponaV2: Driving World Model with Comprehensive Future Reasoning","ref_index":82,"is_internal_anchor":true},{"citing_arxiv_id":"2605.10426","citing_title":"CoWorld-VLA: Thinking in a Multi-Expert World Model for Autonomous Driving","ref_index":53,"is_internal_anchor":true},{"citing_arxiv_id":"2604.00813","citing_title":"DVGT-2: Vision-Geometry-Action Model for Autonomous Driving at Scale","ref_index":81,"is_internal_anchor":true},{"citing_arxiv_id":"2605.12369","citing_title":"GuidedVLA: Specifying Task-Relevant Factors via Plug-and-Play Action Attention Specialization","ref_index":90,"is_internal_anchor":true},{"citing_arxiv_id":"2605.08830","citing_title":"VECTOR-Drive: Tightly Coupled Vision-Language and Trajectory Expert Routing for End-to-End Autonomous Driving","ref_index":26,"is_internal_anchor":true},{"citing_arxiv_id":"2605.10426","citing_title":"CoWorld-VLA: Thinking in a Multi-Expert World Model for Autonomous Driving","ref_index":53,"is_internal_anchor":true},{"citing_arxiv_id":"2604.24514","citing_title":"SceneSelect: Selective Learning for Trajectory Scene Classification and Expert Scheduling","ref_index":37,"is_internal_anchor":true},{"citing_arxiv_id":"2604.19145","citing_title":"ST-Prune: Training-Free Spatio-Temporal Token Pruning for Vision-Language Models in Autonomous Driving","ref_index":30,"is_internal_anchor":true},{"citing_arxiv_id":"2604.09059","citing_title":"Learning Vision-Language-Action World Models for Autonomous Driving","ref_index":72,"is_internal_anchor":true},{"citing_arxiv_id":"2604.08719","citing_title":"LMGenDrive: Bridging Multimodal Understanding and Generative World Modeling for End-to-End Driving","ref_index":61,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OCUAVJ6C2NN4A57DNLXOAHBLGJ","json":"https://pith.science/pith/OCUAVJ6C2NN4A57DNLXOAHBLGJ.json","graph_json":"https://pith.science/api/pith-number/OCUAVJ6C2NN4A57DNLXOAHBLGJ/graph.json","events_json":"https://pith.science/api/pith-number/OCUAVJ6C2NN4A57DNLXOAHBLGJ/events.json","paper":"https://pith.science/paper/OCUAVJ6C"},"agent_actions":{"view_html":"https://pith.science/pith/OCUAVJ6C2NN4A57DNLXOAHBLGJ","download_json":"https://pith.science/pith/OCUAVJ6C2NN4A57DNLXOAHBLGJ.json","view_paper":"https://pith.science/paper/OCUAVJ6C","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.16278&json=true","fetch_graph":"https://pith.science/api/pith-number/OCUAVJ6C2NN4A57DNLXOAHBLGJ/graph.json","fetch_events":"https://pith.science/api/pith-number/OCUAVJ6C2NN4A57DNLXOAHBLGJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OCUAVJ6C2NN4A57DNLXOAHBLGJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OCUAVJ6C2NN4A57DNLXOAHBLGJ/action/storage_attestation","attest_author":"https://pith.science/pith/OCUAVJ6C2NN4A57DNLXOAHBLGJ/action/author_attestation","sign_citation":"https://pith.science/pith/OCUAVJ6C2NN4A57DNLXOAHBLGJ/action/citation_signature","submit_replication":"https://pith.science/pith/OCUAVJ6C2NN4A57DNLXOAHBLGJ/action/replication_record"}},"created_at":"2026-05-20T00:05:28.995922+00:00","updated_at":"2026-05-20T00:05:28.995922+00:00"}