{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:37WOZUNJSULFB7OZWD5AOZC7PU","short_pith_number":"pith:37WOZUNJ","schema_version":"1.0","canonical_sha256":"dfececd1a9951650fdd9b0fa07645f7d13d7e635f0a07df9d16a931570a17ad7","source":{"kind":"arxiv","id":"2605.29028","version":1},"attestation_state":"computed","paper":{"title":"Return-to-Go Is More Than a Number: Q-Guided Alignment for Return-Conditioned Supervised Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Weitong Zhang, Yuxiao Yang","submitted_at":"2026-05-27T19:24:35Z","abstract_excerpt":"Conditioned Sequence Models (CSMs) learn policies by treating return-to-go (RTG) as a control signal. However, existing CSMs often treat the RTGs as simple numerical inputs rather than aligning them with the performance of their policies. In this paper, we propose Q-ALIGN DT, a framework that enforces this alignment by ensuring the $Q$-value of the output policy is consistent with the input RTG. By leveraging a $Q$ function to provide dense guidance to CSMs and further fine-tuning it using an RTG-perturbation technique with the CSM, our method ensures that higher RTGs are consistently mapped t"},"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":"2605.29028","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-27T19:24:35Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"4e265d3d93a65d1a9ea1676c27489efd68f92e68839442be613810ad8a1b4cde","abstract_canon_sha256":"2d1531ab7ed42b9ac52855b4df6f560bb83cf3c9047a4967c004169a5be606c7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T01:04:43.863084Z","signature_b64":"vPQe+5Sw2aHgOUuszhz3eItr/24gJwq0oYqAnANNrIqMrNNDnJ2OxTu6pX8Ep910Luo9qHr+2468M2XHqSEiBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"dfececd1a9951650fdd9b0fa07645f7d13d7e635f0a07df9d16a931570a17ad7","last_reissued_at":"2026-05-29T01:04:43.862629Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T01:04:43.862629Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Return-to-Go Is More Than a Number: Q-Guided Alignment for Return-Conditioned Supervised Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Weitong Zhang, Yuxiao Yang","submitted_at":"2026-05-27T19:24:35Z","abstract_excerpt":"Conditioned Sequence Models (CSMs) learn policies by treating return-to-go (RTG) as a control signal. However, existing CSMs often treat the RTGs as simple numerical inputs rather than aligning them with the performance of their policies. In this paper, we propose Q-ALIGN DT, a framework that enforces this alignment by ensuring the $Q$-value of the output policy is consistent with the input RTG. By leveraging a $Q$ function to provide dense guidance to CSMs and further fine-tuning it using an RTG-perturbation technique with the CSM, our method ensures that higher RTGs are consistently mapped t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.29028","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/2605.29028/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":"2605.29028","created_at":"2026-05-29T01:04:43.862700+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.29028v1","created_at":"2026-05-29T01:04:43.862700+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.29028","created_at":"2026-05-29T01:04:43.862700+00:00"},{"alias_kind":"pith_short_12","alias_value":"37WOZUNJSULF","created_at":"2026-05-29T01:04:43.862700+00:00"},{"alias_kind":"pith_short_16","alias_value":"37WOZUNJSULFB7OZ","created_at":"2026-05-29T01:04:43.862700+00:00"},{"alias_kind":"pith_short_8","alias_value":"37WOZUNJ","created_at":"2026-05-29T01:04:43.862700+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/37WOZUNJSULFB7OZWD5AOZC7PU","json":"https://pith.science/pith/37WOZUNJSULFB7OZWD5AOZC7PU.json","graph_json":"https://pith.science/api/pith-number/37WOZUNJSULFB7OZWD5AOZC7PU/graph.json","events_json":"https://pith.science/api/pith-number/37WOZUNJSULFB7OZWD5AOZC7PU/events.json","paper":"https://pith.science/paper/37WOZUNJ"},"agent_actions":{"view_html":"https://pith.science/pith/37WOZUNJSULFB7OZWD5AOZC7PU","download_json":"https://pith.science/pith/37WOZUNJSULFB7OZWD5AOZC7PU.json","view_paper":"https://pith.science/paper/37WOZUNJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.29028&json=true","fetch_graph":"https://pith.science/api/pith-number/37WOZUNJSULFB7OZWD5AOZC7PU/graph.json","fetch_events":"https://pith.science/api/pith-number/37WOZUNJSULFB7OZWD5AOZC7PU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/37WOZUNJSULFB7OZWD5AOZC7PU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/37WOZUNJSULFB7OZWD5AOZC7PU/action/storage_attestation","attest_author":"https://pith.science/pith/37WOZUNJSULFB7OZWD5AOZC7PU/action/author_attestation","sign_citation":"https://pith.science/pith/37WOZUNJSULFB7OZWD5AOZC7PU/action/citation_signature","submit_replication":"https://pith.science/pith/37WOZUNJSULFB7OZWD5AOZC7PU/action/replication_record"}},"created_at":"2026-05-29T01:04:43.862700+00:00","updated_at":"2026-05-29T01:04:43.862700+00:00"}