{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:YJ7JREO57MDUNDISZEOLIIXFIR","short_pith_number":"pith:YJ7JREO5","schema_version":"1.0","canonical_sha256":"c27e9891ddfb07468d12c91cb422e54464fa0f0a42db9de6c0da02918d8c204e","source":{"kind":"arxiv","id":"2504.11054","version":1},"attestation_state":"computed","paper":{"title":"Zero-Shot Whole-Body Humanoid Control via Behavioral Foundation Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Ahmed Touati, Alessandro Lazaric, Andrea Tirinzoni, Anssi Kanervisto, Jesse Farebrother, Mateusz Guzek, Matteo Pirotta, Yingchen Xu","submitted_at":"2025-04-15T10:41:11Z","abstract_excerpt":"Unsupervised reinforcement learning (RL) aims at pre-training agents that can solve a wide range of downstream tasks in complex environments. Despite recent advancements, existing approaches suffer from several limitations: they may require running an RL process on each downstream task to achieve a satisfactory performance, they may need access to datasets with good coverage or well-curated task-specific samples, or they may pre-train policies with unsupervised losses that are poorly correlated with the downstream tasks of interest. In this paper, we introduce a novel algorithm regularizing un"},"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":"2504.11054","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-04-15T10:41:11Z","cross_cats_sorted":[],"title_canon_sha256":"6939a091bd644494940d5ef18a8035aabb9aba1804006de3edf72cd2597737cc","abstract_canon_sha256":"006b2fd3d1ae024ac207398ae3bdc67732a2bb9ef0c3c002ad4ce53ac52f96c1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:49:28.324146Z","signature_b64":"RMlgeqT/MmYwp7DFyRZdTYWdV3ZUGWMZ6EeprHwnkMWbDQZ2TL4O4wvGH+x4Vd95A2T21E0OsrVARDLXk441Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c27e9891ddfb07468d12c91cb422e54464fa0f0a42db9de6c0da02918d8c204e","last_reissued_at":"2026-07-05T10:49:28.323660Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:49:28.323660Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Zero-Shot Whole-Body Humanoid Control via Behavioral Foundation Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Ahmed Touati, Alessandro Lazaric, Andrea Tirinzoni, Anssi Kanervisto, Jesse Farebrother, Mateusz Guzek, Matteo Pirotta, Yingchen Xu","submitted_at":"2025-04-15T10:41:11Z","abstract_excerpt":"Unsupervised reinforcement learning (RL) aims at pre-training agents that can solve a wide range of downstream tasks in complex environments. Despite recent advancements, existing approaches suffer from several limitations: they may require running an RL process on each downstream task to achieve a satisfactory performance, they may need access to datasets with good coverage or well-curated task-specific samples, or they may pre-train policies with unsupervised losses that are poorly correlated with the downstream tasks of interest. In this paper, we introduce a novel algorithm regularizing un"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.11054","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/2504.11054/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":"2504.11054","created_at":"2026-07-05T10:49:28.323719+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.11054v1","created_at":"2026-07-05T10:49:28.323719+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.11054","created_at":"2026-07-05T10:49:28.323719+00:00"},{"alias_kind":"pith_short_12","alias_value":"YJ7JREO57MDU","created_at":"2026-07-05T10:49:28.323719+00:00"},{"alias_kind":"pith_short_16","alias_value":"YJ7JREO57MDUNDIS","created_at":"2026-07-05T10:49:28.323719+00:00"},{"alias_kind":"pith_short_8","alias_value":"YJ7JREO5","created_at":"2026-07-05T10:49:28.323719+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.21271","citing_title":"Reward-free Pretraining for Reinforcement Learning via Occupancy Coverage Maximization","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2606.20705","citing_title":"MotionPyramid: Hierarchical Motion Representation and Residual Interfaces","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2606.10340","citing_title":"OMG: Omni-Modal Motion Generation for Generalist Humanoid Control","ref_index":67,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29980","citing_title":"Exploration and Online Transfer with Behavioral Foundation Models","ref_index":65,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23551","citing_title":"Goal-Conditioned Agents that Learn Everything All at Once","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21707","citing_title":"Zero-shot adaptation to order book dynamics","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YJ7JREO57MDUNDISZEOLIIXFIR","json":"https://pith.science/pith/YJ7JREO57MDUNDISZEOLIIXFIR.json","graph_json":"https://pith.science/api/pith-number/YJ7JREO57MDUNDISZEOLIIXFIR/graph.json","events_json":"https://pith.science/api/pith-number/YJ7JREO57MDUNDISZEOLIIXFIR/events.json","paper":"https://pith.science/paper/YJ7JREO5"},"agent_actions":{"view_html":"https://pith.science/pith/YJ7JREO57MDUNDISZEOLIIXFIR","download_json":"https://pith.science/pith/YJ7JREO57MDUNDISZEOLIIXFIR.json","view_paper":"https://pith.science/paper/YJ7JREO5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.11054&json=true","fetch_graph":"https://pith.science/api/pith-number/YJ7JREO57MDUNDISZEOLIIXFIR/graph.json","fetch_events":"https://pith.science/api/pith-number/YJ7JREO57MDUNDISZEOLIIXFIR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YJ7JREO57MDUNDISZEOLIIXFIR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YJ7JREO57MDUNDISZEOLIIXFIR/action/storage_attestation","attest_author":"https://pith.science/pith/YJ7JREO57MDUNDISZEOLIIXFIR/action/author_attestation","sign_citation":"https://pith.science/pith/YJ7JREO57MDUNDISZEOLIIXFIR/action/citation_signature","submit_replication":"https://pith.science/pith/YJ7JREO57MDUNDISZEOLIIXFIR/action/replication_record"}},"created_at":"2026-07-05T10:49:28.323719+00:00","updated_at":"2026-07-05T10:49:28.323719+00:00"}