{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:MS7VSGNKJYHPGFSLQKYLQFKVBE","short_pith_number":"pith:MS7VSGNK","schema_version":"1.0","canonical_sha256":"64bf5919aa4e0ef3164b82b0b81555093028f9269cacb748d5565ac60ada404a","source":{"kind":"arxiv","id":"2206.11251","version":2},"attestation_state":"computed","paper":{"title":"Behavior Transformers: Cloning $k$ modes with one stone","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.RO"],"primary_cat":"cs.LG","authors_text":"Ariuntuya Altanzaya, Lerrel Pinto, Nur Muhammad Mahi Shafiullah, Zichen Jeff Cui","submitted_at":"2022-06-22T17:57:08Z","abstract_excerpt":"While behavior learning has made impressive progress in recent times, it lags behind computer vision and natural language processing due to its inability to leverage large, human-generated datasets. Human behaviors have wide variance, multiple modes, and human demonstrations typically do not come with reward labels. These properties limit the applicability of current methods in Offline RL and Behavioral Cloning to learn from large, pre-collected datasets. In this work, we present Behavior Transformer (BeT), a new technique to model unlabeled demonstration data with multiple modes. BeT retrofit"},"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":"2206.11251","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-06-22T17:57:08Z","cross_cats_sorted":["cs.AI","cs.CV","cs.RO"],"title_canon_sha256":"0cadc06268f24628327ebf606d06e4d26e5bcbe03f9283c8e0271b8b224ec0f3","abstract_canon_sha256":"4e09a6ad8bc7d4393d3b65d0bc1f850f4392537d6494370f107828282842cc80"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:05:48.865721Z","signature_b64":"wOJNf7LtV1+wuDeXCL37bHdU4Cjo/jUxVQEVK5LGtsBHRI8ePcotS3KspWo4QyF8X18ZPeh0/YObAXemgvIaCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"64bf5919aa4e0ef3164b82b0b81555093028f9269cacb748d5565ac60ada404a","last_reissued_at":"2026-07-05T05:05:48.865211Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:05:48.865211Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Behavior Transformers: Cloning $k$ modes with one stone","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.RO"],"primary_cat":"cs.LG","authors_text":"Ariuntuya Altanzaya, Lerrel Pinto, Nur Muhammad Mahi Shafiullah, Zichen Jeff Cui","submitted_at":"2022-06-22T17:57:08Z","abstract_excerpt":"While behavior learning has made impressive progress in recent times, it lags behind computer vision and natural language processing due to its inability to leverage large, human-generated datasets. Human behaviors have wide variance, multiple modes, and human demonstrations typically do not come with reward labels. These properties limit the applicability of current methods in Offline RL and Behavioral Cloning to learn from large, pre-collected datasets. In this work, we present Behavior Transformer (BeT), a new technique to model unlabeled demonstration data with multiple modes. BeT retrofit"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2206.11251","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/2206.11251/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":"2206.11251","created_at":"2026-07-05T05:05:48.865273+00:00"},{"alias_kind":"arxiv_version","alias_value":"2206.11251v2","created_at":"2026-07-05T05:05:48.865273+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.11251","created_at":"2026-07-05T05:05:48.865273+00:00"},{"alias_kind":"pith_short_12","alias_value":"MS7VSGNKJYHP","created_at":"2026-07-05T05:05:48.865273+00:00"},{"alias_kind":"pith_short_16","alias_value":"MS7VSGNKJYHPGFSL","created_at":"2026-07-05T05:05:48.865273+00:00"},{"alias_kind":"pith_short_8","alias_value":"MS7VSGNK","created_at":"2026-07-05T05:05:48.865273+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":13,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.00229","citing_title":"Continuous Reasoning for Vision-Language-Action","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00089","citing_title":"Can Predicted Dynamics Exist in the Physical World?","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25537","citing_title":"Action-Prior Denoising for Smooth Real-Time Chunking","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.29864","citing_title":"LLM-Guided Future Hypotheses for Horizon-Aware Exploration in Multi-Step Robot Manipulation","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14598","citing_title":"DSSP: Diffusion State Space Policy with Full-History Encoding","ref_index":46,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15536","citing_title":"SkiP: When to Skip and When to Refine for Efficient Robot Manipulation","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2510.01433","citing_title":"AFFORD2ACT: Affordance-Guided Automatic Keypoint Selection for Generalizable and Lightweight Robotic Manipulation","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2208.06193","citing_title":"Diffusion Policies as an Expressive Policy Class for Offline Reinforcement Learning","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2401.02117","citing_title":"Mobile ALOHA: Learning Bimanual Mobile Manipulation with Low-Cost Whole-Body Teleoperation","ref_index":77,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13428","citing_title":"SID: Sliding into Distribution for Robust Few-Demonstration Manipulation","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2504.02792","citing_title":"Unified World Models: Coupling Video and Action Diffusion for Pretraining on Large Robotic Datasets","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2304.13705","citing_title":"Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2604.05544","citing_title":"Referring-Aware Visuomotor Policy Learning for Closed-Loop Manipulation","ref_index":31,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MS7VSGNKJYHPGFSLQKYLQFKVBE","json":"https://pith.science/pith/MS7VSGNKJYHPGFSLQKYLQFKVBE.json","graph_json":"https://pith.science/api/pith-number/MS7VSGNKJYHPGFSLQKYLQFKVBE/graph.json","events_json":"https://pith.science/api/pith-number/MS7VSGNKJYHPGFSLQKYLQFKVBE/events.json","paper":"https://pith.science/paper/MS7VSGNK"},"agent_actions":{"view_html":"https://pith.science/pith/MS7VSGNKJYHPGFSLQKYLQFKVBE","download_json":"https://pith.science/pith/MS7VSGNKJYHPGFSLQKYLQFKVBE.json","view_paper":"https://pith.science/paper/MS7VSGNK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2206.11251&json=true","fetch_graph":"https://pith.science/api/pith-number/MS7VSGNKJYHPGFSLQKYLQFKVBE/graph.json","fetch_events":"https://pith.science/api/pith-number/MS7VSGNKJYHPGFSLQKYLQFKVBE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MS7VSGNKJYHPGFSLQKYLQFKVBE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MS7VSGNKJYHPGFSLQKYLQFKVBE/action/storage_attestation","attest_author":"https://pith.science/pith/MS7VSGNKJYHPGFSLQKYLQFKVBE/action/author_attestation","sign_citation":"https://pith.science/pith/MS7VSGNKJYHPGFSLQKYLQFKVBE/action/citation_signature","submit_replication":"https://pith.science/pith/MS7VSGNKJYHPGFSLQKYLQFKVBE/action/replication_record"}},"created_at":"2026-07-05T05:05:48.865273+00:00","updated_at":"2026-07-05T05:05:48.865273+00:00"}