{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:NK52BBK32INMCS2JXKV2LKHM42","short_pith_number":"pith:NK52BBK3","schema_version":"1.0","canonical_sha256":"6abba0855bd21ac14b49baaba5a8ece6b70fe303f4a53320fa205b359740d593","source":{"kind":"arxiv","id":"2502.05454","version":2},"attestation_state":"computed","paper":{"title":"Temporal Representation Alignment: Successor Features Enable Emergent Compositionality in Robot Instruction Following","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.RO","authors_text":"Anca Dragan, Bill Chunyuan Zheng, Kuan Fang, Sergey Levine, Vivek Myers","submitted_at":"2025-02-08T05:26:29Z","abstract_excerpt":"Effective task representations should facilitate compositionality, such that after learning a variety of basic tasks, an agent can perform compound tasks consisting of multiple steps simply by composing the representations of the constituent steps together. While this is conceptually simple and appealing, it is not clear how to automatically learn representations that enable this sort of compositionality. We show that learning to associate the representations of current and future states with a temporal alignment loss can improve compositional generalization, even in the absence of any explici"},"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":"2502.05454","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2025-02-08T05:26:29Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"80624f066f375e61ab96cd0e5fc7da2abb69af2a9f2536554c3d87939d4e16c4","abstract_canon_sha256":"7d6ae18a5bf99abbbbc8d53ebac1eb15a98928b2cda2f21289c127f0096da79c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:13:47.798188Z","signature_b64":"HlAc05h0c/vIq1FzatiHGIFd1wQFAd8TNuBW6jOYbUxrV3Hs0L+MMxsvcMCXhjRt3mfkvyQKvAwotxejxGWTCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6abba0855bd21ac14b49baaba5a8ece6b70fe303f4a53320fa205b359740d593","last_reissued_at":"2026-07-05T10:13:47.797699Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:13:47.797699Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Temporal Representation Alignment: Successor Features Enable Emergent Compositionality in Robot Instruction Following","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.RO","authors_text":"Anca Dragan, Bill Chunyuan Zheng, Kuan Fang, Sergey Levine, Vivek Myers","submitted_at":"2025-02-08T05:26:29Z","abstract_excerpt":"Effective task representations should facilitate compositionality, such that after learning a variety of basic tasks, an agent can perform compound tasks consisting of multiple steps simply by composing the representations of the constituent steps together. While this is conceptually simple and appealing, it is not clear how to automatically learn representations that enable this sort of compositionality. We show that learning to associate the representations of current and future states with a temporal alignment loss can improve compositional generalization, even in the absence of any explici"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.05454","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/2502.05454/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":"2502.05454","created_at":"2026-07-05T10:13:47.797758+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.05454v2","created_at":"2026-07-05T10:13:47.797758+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.05454","created_at":"2026-07-05T10:13:47.797758+00:00"},{"alias_kind":"pith_short_12","alias_value":"NK52BBK32INM","created_at":"2026-07-05T10:13:47.797758+00:00"},{"alias_kind":"pith_short_16","alias_value":"NK52BBK32INMCS2J","created_at":"2026-07-05T10:13:47.797758+00:00"},{"alias_kind":"pith_short_8","alias_value":"NK52BBK3","created_at":"2026-07-05T10:13:47.797758+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2506.10137","citing_title":"Self-Predictive Representations for Combinatorial Generalization in Behavioral Cloning","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09364","citing_title":"Multi-scale Predictive Representations for Goal-conditioned Reinforcement Learning","ref_index":29,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NK52BBK32INMCS2JXKV2LKHM42","json":"https://pith.science/pith/NK52BBK32INMCS2JXKV2LKHM42.json","graph_json":"https://pith.science/api/pith-number/NK52BBK32INMCS2JXKV2LKHM42/graph.json","events_json":"https://pith.science/api/pith-number/NK52BBK32INMCS2JXKV2LKHM42/events.json","paper":"https://pith.science/paper/NK52BBK3"},"agent_actions":{"view_html":"https://pith.science/pith/NK52BBK32INMCS2JXKV2LKHM42","download_json":"https://pith.science/pith/NK52BBK32INMCS2JXKV2LKHM42.json","view_paper":"https://pith.science/paper/NK52BBK3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.05454&json=true","fetch_graph":"https://pith.science/api/pith-number/NK52BBK32INMCS2JXKV2LKHM42/graph.json","fetch_events":"https://pith.science/api/pith-number/NK52BBK32INMCS2JXKV2LKHM42/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NK52BBK32INMCS2JXKV2LKHM42/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NK52BBK32INMCS2JXKV2LKHM42/action/storage_attestation","attest_author":"https://pith.science/pith/NK52BBK32INMCS2JXKV2LKHM42/action/author_attestation","sign_citation":"https://pith.science/pith/NK52BBK32INMCS2JXKV2LKHM42/action/citation_signature","submit_replication":"https://pith.science/pith/NK52BBK32INMCS2JXKV2LKHM42/action/replication_record"}},"created_at":"2026-07-05T10:13:47.797758+00:00","updated_at":"2026-07-05T10:13:47.797758+00:00"}