{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:TBSADBWD354MP7KMMQ63B6ILXF","short_pith_number":"pith:TBSADBWD","schema_version":"1.0","canonical_sha256":"98640186c3df78c7fd4c643db0f90bb965d53e7b83097060b2e70bb105cbcba8","source":{"kind":"arxiv","id":"2306.03346","version":3},"attestation_state":"computed","paper":{"title":"Stabilizing Contrastive RL: Techniques for Robotic Goal Reaching from Offline Data","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Benjamin Eysenbach, Chongyi Zheng, Homer Walke, Kuan Fang, Patrick Yin, Ruslan Salakhutdinov, Sergey Levine","submitted_at":"2023-06-06T01:36:56Z","abstract_excerpt":"Robotic systems that rely primarily on self-supervised learning have the potential to decrease the amount of human annotation and engineering effort required to learn control strategies. In the same way that prior robotic systems have leveraged self-supervised techniques from computer vision (CV) and natural language processing (NLP), our work builds on prior work showing that the reinforcement learning (RL) itself can be cast as a self-supervised problem: learning to reach any goal without human-specified rewards or labels. Despite the seeming appeal, little (if any) prior work has demonstrat"},"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":"2306.03346","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-06-06T01:36:56Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"5b60e1a7d5e0f45aa0fab2ef64bc6f79a91a7ee6a16b86dbfd9b6ae64e41b815","abstract_canon_sha256":"97b585bcadd8b843a9b16d56c5dd90277d3f0c44e57309b9638eb0ad805f15c0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:18:38.250824Z","signature_b64":"S+S8KqncHQh7ND49SDSNFhVLxZ9bnz1d2FMN9an8E9VMJABGf2NliF3cjuhUQom2KpPSOY9hXIUyti+3Nzi2Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"98640186c3df78c7fd4c643db0f90bb965d53e7b83097060b2e70bb105cbcba8","last_reissued_at":"2026-07-05T11:18:38.250347Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:18:38.250347Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Stabilizing Contrastive RL: Techniques for Robotic Goal Reaching from Offline Data","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Benjamin Eysenbach, Chongyi Zheng, Homer Walke, Kuan Fang, Patrick Yin, Ruslan Salakhutdinov, Sergey Levine","submitted_at":"2023-06-06T01:36:56Z","abstract_excerpt":"Robotic systems that rely primarily on self-supervised learning have the potential to decrease the amount of human annotation and engineering effort required to learn control strategies. In the same way that prior robotic systems have leveraged self-supervised techniques from computer vision (CV) and natural language processing (NLP), our work builds on prior work showing that the reinforcement learning (RL) itself can be cast as a self-supervised problem: learning to reach any goal without human-specified rewards or labels. Despite the seeming appeal, little (if any) prior work has demonstrat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.03346","kind":"arxiv","version":3},"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/2306.03346/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":"2306.03346","created_at":"2026-07-05T11:18:38.250406+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.03346v3","created_at":"2026-07-05T11:18:38.250406+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.03346","created_at":"2026-07-05T11:18:38.250406+00:00"},{"alias_kind":"pith_short_12","alias_value":"TBSADBWD354M","created_at":"2026-07-05T11:18:38.250406+00:00"},{"alias_kind":"pith_short_16","alias_value":"TBSADBWD354MP7KM","created_at":"2026-07-05T11:18:38.250406+00:00"},{"alias_kind":"pith_short_8","alias_value":"TBSADBWD","created_at":"2026-07-05T11:18:38.250406+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.20867","citing_title":"FOCA: Future-Oriented Conditioning for Data-Efficient Vision-Language-Action Adaptation","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08512","citing_title":"MoMo: Conditioned Contrastive Representation Learning for Preference-Modulated Planning","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13554","citing_title":"Self-Supervised On-Policy Reinforcement Learning via Contrastive Proximal Policy Optimisation","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08512","citing_title":"MoMo: Conditioned Contrastive Representation Learning for Preference-Modulated Planning","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01862","citing_title":"QHyer: Q-conditioned Hybrid Attention-mamba Transformer for Offline Goal-conditioned RL","ref_index":230,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TBSADBWD354MP7KMMQ63B6ILXF","json":"https://pith.science/pith/TBSADBWD354MP7KMMQ63B6ILXF.json","graph_json":"https://pith.science/api/pith-number/TBSADBWD354MP7KMMQ63B6ILXF/graph.json","events_json":"https://pith.science/api/pith-number/TBSADBWD354MP7KMMQ63B6ILXF/events.json","paper":"https://pith.science/paper/TBSADBWD"},"agent_actions":{"view_html":"https://pith.science/pith/TBSADBWD354MP7KMMQ63B6ILXF","download_json":"https://pith.science/pith/TBSADBWD354MP7KMMQ63B6ILXF.json","view_paper":"https://pith.science/paper/TBSADBWD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.03346&json=true","fetch_graph":"https://pith.science/api/pith-number/TBSADBWD354MP7KMMQ63B6ILXF/graph.json","fetch_events":"https://pith.science/api/pith-number/TBSADBWD354MP7KMMQ63B6ILXF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TBSADBWD354MP7KMMQ63B6ILXF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TBSADBWD354MP7KMMQ63B6ILXF/action/storage_attestation","attest_author":"https://pith.science/pith/TBSADBWD354MP7KMMQ63B6ILXF/action/author_attestation","sign_citation":"https://pith.science/pith/TBSADBWD354MP7KMMQ63B6ILXF/action/citation_signature","submit_replication":"https://pith.science/pith/TBSADBWD354MP7KMMQ63B6ILXF/action/replication_record"}},"created_at":"2026-07-05T11:18:38.250406+00:00","updated_at":"2026-07-05T11:18:38.250406+00:00"}