{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:D34DVYO2EMGL6OY3GZGEUINRF2","short_pith_number":"pith:D34DVYO2","schema_version":"1.0","canonical_sha256":"1ef83ae1da230cbf3b1b364c4a21b12e823795b9dd6ea52ee7625824a09455f5","source":{"kind":"arxiv","id":"2506.11470","version":1},"attestation_state":"computed","paper":{"title":"Multi-Loco: Unifying Multi-Embodiment Legged Locomotion via Reinforcement Learning Augmented Diffusion","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Guo Junde, Hua Chen, Patrick Wensing, Shunpeng Yang, Wei Zhang, Zhefeng Cao, Zhen Fu","submitted_at":"2025-06-13T05:16:30Z","abstract_excerpt":"Generalizing locomotion policies across diverse legged robots with varying morphologies is a key challenge due to differences in observation/action dimensions and system dynamics. In this work, we propose Multi-Loco, a novel unified framework combining a morphology-agnostic generative diffusion model with a lightweight residual policy optimized via reinforcement learning (RL). The diffusion model captures morphology-invariant locomotion patterns from diverse cross-embodiment datasets, improving generalization and robustness. The residual policy is shared across all embodiments and refines the "},"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":"2506.11470","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2025-06-13T05:16:30Z","cross_cats_sorted":[],"title_canon_sha256":"cdc8413f309980eeb6b22ba18db34f2d8731a4950723b3132ecae552886cf484","abstract_canon_sha256":"54c7c49074e3fa0c0eb8ee676c1dd972a729eeb02670f694fc7a224f6edaa992"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:21:05.114342Z","signature_b64":"ChqHNRCfPH7gql0L7ZU5bI+e1iIaX0pEAm7ykwyZHdGJLVLgQ8HkGqDjVA9SY7S14in4yqL7TEXfqeWhlJ2mBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1ef83ae1da230cbf3b1b364c4a21b12e823795b9dd6ea52ee7625824a09455f5","last_reissued_at":"2026-07-05T11:21:05.113695Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:21:05.113695Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multi-Loco: Unifying Multi-Embodiment Legged Locomotion via Reinforcement Learning Augmented Diffusion","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Guo Junde, Hua Chen, Patrick Wensing, Shunpeng Yang, Wei Zhang, Zhefeng Cao, Zhen Fu","submitted_at":"2025-06-13T05:16:30Z","abstract_excerpt":"Generalizing locomotion policies across diverse legged robots with varying morphologies is a key challenge due to differences in observation/action dimensions and system dynamics. In this work, we propose Multi-Loco, a novel unified framework combining a morphology-agnostic generative diffusion model with a lightweight residual policy optimized via reinforcement learning (RL). The diffusion model captures morphology-invariant locomotion patterns from diverse cross-embodiment datasets, improving generalization and robustness. The residual policy is shared across all embodiments and refines the "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.11470","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/2506.11470/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":"2506.11470","created_at":"2026-07-05T11:21:05.113753+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.11470v1","created_at":"2026-07-05T11:21:05.113753+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.11470","created_at":"2026-07-05T11:21:05.113753+00:00"},{"alias_kind":"pith_short_12","alias_value":"D34DVYO2EMGL","created_at":"2026-07-05T11:21:05.113753+00:00"},{"alias_kind":"pith_short_16","alias_value":"D34DVYO2EMGL6OY3","created_at":"2026-07-05T11:21:05.113753+00:00"},{"alias_kind":"pith_short_8","alias_value":"D34DVYO2","created_at":"2026-07-05T11:21:05.113753+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24089","citing_title":"DynaWM: Dynamics-Aware Distillation with World Model and Momentum Targets for Smooth Locomotion over Continuous Stairs","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2606.07424","citing_title":"Rapid co-design of Buoyancy-assisted robots for Challenging Locomotion using Gaussian Evolutionary Specialists","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23733","citing_title":"Any2Any: Efficient Cross-Embodiment Transfer for Humanoid Whole-Body Tracking","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23733","citing_title":"Any2Any: Efficient Cross-Embodiment Transfer for Humanoid Whole-Body Tracking","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2510.27420","citing_title":"Towards a Multi-Embodied Grasping Agent","ref_index":51,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/D34DVYO2EMGL6OY3GZGEUINRF2","json":"https://pith.science/pith/D34DVYO2EMGL6OY3GZGEUINRF2.json","graph_json":"https://pith.science/api/pith-number/D34DVYO2EMGL6OY3GZGEUINRF2/graph.json","events_json":"https://pith.science/api/pith-number/D34DVYO2EMGL6OY3GZGEUINRF2/events.json","paper":"https://pith.science/paper/D34DVYO2"},"agent_actions":{"view_html":"https://pith.science/pith/D34DVYO2EMGL6OY3GZGEUINRF2","download_json":"https://pith.science/pith/D34DVYO2EMGL6OY3GZGEUINRF2.json","view_paper":"https://pith.science/paper/D34DVYO2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.11470&json=true","fetch_graph":"https://pith.science/api/pith-number/D34DVYO2EMGL6OY3GZGEUINRF2/graph.json","fetch_events":"https://pith.science/api/pith-number/D34DVYO2EMGL6OY3GZGEUINRF2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/D34DVYO2EMGL6OY3GZGEUINRF2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/D34DVYO2EMGL6OY3GZGEUINRF2/action/storage_attestation","attest_author":"https://pith.science/pith/D34DVYO2EMGL6OY3GZGEUINRF2/action/author_attestation","sign_citation":"https://pith.science/pith/D34DVYO2EMGL6OY3GZGEUINRF2/action/citation_signature","submit_replication":"https://pith.science/pith/D34DVYO2EMGL6OY3GZGEUINRF2/action/replication_record"}},"created_at":"2026-07-05T11:21:05.113753+00:00","updated_at":"2026-07-05T11:21:05.113753+00:00"}