{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:QVV4CYL5PRPHJVZTAWPFZDXJDP","short_pith_number":"pith:QVV4CYL5","schema_version":"1.0","canonical_sha256":"856bc1617d7c5e74d733059e5c8ee91bd007aed182e048d455a95377e2d3a4c6","source":{"kind":"arxiv","id":"2605.30795","version":1},"attestation_state":"computed","paper":{"title":"Feat2Go: Visual Feature-Grounded Value Estimation for Embodied Reinforcement Learning","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Bingqing Wei, Junyang Shu, Yongtao Wang, Zhiwei Lin","submitted_at":"2026-05-29T03:36:33Z","abstract_excerpt":"Reinforcement learning is a promising approach for improving the capabilities of vision-language-action (VLA) models while avoiding the heavy data requirements of imitation learning. However, its effectiveness for VLA models is often constrained by sparse supervision and the difficulty of designing informative reward signals for long-horizon manipulation. In this work, we present Feat2Go, a fine-grained value estimation framework for embodied reinforcement learning. Specifically, Feat2Go first derives a continuous progress target from a pretrained visual world model by measuring patch-level si"},"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":"2605.30795","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.RO","submitted_at":"2026-05-29T03:36:33Z","cross_cats_sorted":[],"title_canon_sha256":"cc691c86e7f79213f54ecb19b806c8c23c9deb68b81a8d457bd569134eaa4475","abstract_canon_sha256":"9395cd26bacd93c5323dfad6977c88cc0c4fb7fce9bcf1f4542c974500c4747f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-01T01:03:17.256369Z","signature_b64":"mERy2dHnA0Bvb/PrjhSrRAxu/TTvkoyO+jydBLO7Giml1tZBn7bCyiFa6xcaBxhRDXDkez21pGKNsrCGIKPSCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"856bc1617d7c5e74d733059e5c8ee91bd007aed182e048d455a95377e2d3a4c6","last_reissued_at":"2026-06-01T01:03:17.255751Z","signature_status":"signed_v1","first_computed_at":"2026-06-01T01:03:17.255751Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Feat2Go: Visual Feature-Grounded Value Estimation for Embodied Reinforcement Learning","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Bingqing Wei, Junyang Shu, Yongtao Wang, Zhiwei Lin","submitted_at":"2026-05-29T03:36:33Z","abstract_excerpt":"Reinforcement learning is a promising approach for improving the capabilities of vision-language-action (VLA) models while avoiding the heavy data requirements of imitation learning. However, its effectiveness for VLA models is often constrained by sparse supervision and the difficulty of designing informative reward signals for long-horizon manipulation. In this work, we present Feat2Go, a fine-grained value estimation framework for embodied reinforcement learning. Specifically, Feat2Go first derives a continuous progress target from a pretrained visual world model by measuring patch-level si"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.30795","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/2605.30795/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":"2605.30795","created_at":"2026-06-01T01:03:17.255849+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.30795v1","created_at":"2026-06-01T01:03:17.255849+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.30795","created_at":"2026-06-01T01:03:17.255849+00:00"},{"alias_kind":"pith_short_12","alias_value":"QVV4CYL5PRPH","created_at":"2026-06-01T01:03:17.255849+00:00"},{"alias_kind":"pith_short_16","alias_value":"QVV4CYL5PRPHJVZT","created_at":"2026-06-01T01:03:17.255849+00:00"},{"alias_kind":"pith_short_8","alias_value":"QVV4CYL5","created_at":"2026-06-01T01:03:17.255849+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/QVV4CYL5PRPHJVZTAWPFZDXJDP","json":"https://pith.science/pith/QVV4CYL5PRPHJVZTAWPFZDXJDP.json","graph_json":"https://pith.science/api/pith-number/QVV4CYL5PRPHJVZTAWPFZDXJDP/graph.json","events_json":"https://pith.science/api/pith-number/QVV4CYL5PRPHJVZTAWPFZDXJDP/events.json","paper":"https://pith.science/paper/QVV4CYL5"},"agent_actions":{"view_html":"https://pith.science/pith/QVV4CYL5PRPHJVZTAWPFZDXJDP","download_json":"https://pith.science/pith/QVV4CYL5PRPHJVZTAWPFZDXJDP.json","view_paper":"https://pith.science/paper/QVV4CYL5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.30795&json=true","fetch_graph":"https://pith.science/api/pith-number/QVV4CYL5PRPHJVZTAWPFZDXJDP/graph.json","fetch_events":"https://pith.science/api/pith-number/QVV4CYL5PRPHJVZTAWPFZDXJDP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QVV4CYL5PRPHJVZTAWPFZDXJDP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QVV4CYL5PRPHJVZTAWPFZDXJDP/action/storage_attestation","attest_author":"https://pith.science/pith/QVV4CYL5PRPHJVZTAWPFZDXJDP/action/author_attestation","sign_citation":"https://pith.science/pith/QVV4CYL5PRPHJVZTAWPFZDXJDP/action/citation_signature","submit_replication":"https://pith.science/pith/QVV4CYL5PRPHJVZTAWPFZDXJDP/action/replication_record"}},"created_at":"2026-06-01T01:03:17.255849+00:00","updated_at":"2026-06-01T01:03:17.255849+00:00"}