{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:GRKIDKVF5P4UES25Y5ZVVDLTER","short_pith_number":"pith:GRKIDKVF","schema_version":"1.0","canonical_sha256":"345481aaa5ebf9424b5dc7735a8d73245da113c11c18bd237000b9e9daedb0cb","source":{"kind":"arxiv","id":"2509.21576","version":2},"attestation_state":"computed","paper":{"title":"Vision Language Models Cannot Plan, but Can They Formalize?","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bill Cai, Feng Liu, Jiani Huang, Lifeng Zhou, Li Zhang, Muyu He, Yuchen Liu, Yuxi Zheng, Zijian An, Ziyang Li","submitted_at":"2025-09-25T20:55:11Z","abstract_excerpt":"The advancement of vision language models (VLMs) has empowered embodied agents to accomplish simple multimodal planning tasks, but not long-horizon ones requiring long sequences of actions. In text-only simulations, long-horizon planning has seen significant improvement brought by repositioning the role of LLMs. Instead of directly generating action sequences, LLMs translate the planning domain and problem into a formal planning language like the Planning Domain Definition Language (PDDL), which can call a formal solver to derive the plan in a verifiable manner. In multimodal environments, res"},"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":"2509.21576","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-25T20:55:11Z","cross_cats_sorted":[],"title_canon_sha256":"f3ee2bfecbebfaea2bd99ed56035677ccc2265926e293ebbd69538a40bb28968","abstract_canon_sha256":"2bda5c8f823f7bf0a71219d2223dcd356a8e31ef2e056ecc5dc9497f36758985"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-18T01:14:01.250281Z","signature_b64":"dfEICyn8u9r1IZRUWUeB9k1+0U6QSoxe2vtHHSc9+2KGDXLmIOoyIcLM501ILTFKBuZHh/R8TUvtDnuj8DKUAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"345481aaa5ebf9424b5dc7735a8d73245da113c11c18bd237000b9e9daedb0cb","last_reissued_at":"2026-08-18T01:14:01.248323Z","signature_status":"signed_v1","first_computed_at":"2026-08-18T01:14:01.248323Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Vision Language Models Cannot Plan, but Can They Formalize?","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bill Cai, Feng Liu, Jiani Huang, Lifeng Zhou, Li Zhang, Muyu He, Yuchen Liu, Yuxi Zheng, Zijian An, Ziyang Li","submitted_at":"2025-09-25T20:55:11Z","abstract_excerpt":"The advancement of vision language models (VLMs) has empowered embodied agents to accomplish simple multimodal planning tasks, but not long-horizon ones requiring long sequences of actions. In text-only simulations, long-horizon planning has seen significant improvement brought by repositioning the role of LLMs. Instead of directly generating action sequences, LLMs translate the planning domain and problem into a formal planning language like the Planning Domain Definition Language (PDDL), which can call a formal solver to derive the plan in a verifiable manner. In multimodal environments, res"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.21576","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/2509.21576/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":"2509.21576","created_at":"2026-08-18T01:14:01.249135+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.21576v2","created_at":"2026-08-18T01:14:01.249135+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.21576","created_at":"2026-08-18T01:14:01.249135+00:00"},{"alias_kind":"pith_short_12","alias_value":"GRKIDKVF5P4U","created_at":"2026-08-18T01:14:01.249135+00:00"},{"alias_kind":"pith_short_16","alias_value":"GRKIDKVF5P4UES25","created_at":"2026-08-18T01:14:01.249135+00:00"},{"alias_kind":"pith_short_8","alias_value":"GRKIDKVF","created_at":"2026-08-18T01:14:01.249135+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.13775","citing_title":"RoboEvolve: Co-Evolving Planner-Simulator for Robotic Manipulation with Limited Data","ref_index":58,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GRKIDKVF5P4UES25Y5ZVVDLTER","json":"https://pith.science/pith/GRKIDKVF5P4UES25Y5ZVVDLTER.json","graph_json":"https://pith.science/api/pith-number/GRKIDKVF5P4UES25Y5ZVVDLTER/graph.json","events_json":"https://pith.science/api/pith-number/GRKIDKVF5P4UES25Y5ZVVDLTER/events.json","paper":"https://pith.science/paper/GRKIDKVF"},"agent_actions":{"view_html":"https://pith.science/pith/GRKIDKVF5P4UES25Y5ZVVDLTER","download_json":"https://pith.science/pith/GRKIDKVF5P4UES25Y5ZVVDLTER.json","view_paper":"https://pith.science/paper/GRKIDKVF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.21576&json=true","fetch_graph":"https://pith.science/api/pith-number/GRKIDKVF5P4UES25Y5ZVVDLTER/graph.json","fetch_events":"https://pith.science/api/pith-number/GRKIDKVF5P4UES25Y5ZVVDLTER/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GRKIDKVF5P4UES25Y5ZVVDLTER/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GRKIDKVF5P4UES25Y5ZVVDLTER/action/storage_attestation","attest_author":"https://pith.science/pith/GRKIDKVF5P4UES25Y5ZVVDLTER/action/author_attestation","sign_citation":"https://pith.science/pith/GRKIDKVF5P4UES25Y5ZVVDLTER/action/citation_signature","submit_replication":"https://pith.science/pith/GRKIDKVF5P4UES25Y5ZVVDLTER/action/replication_record"}},"created_at":"2026-08-18T01:14:01.249135+00:00","updated_at":"2026-08-18T01:14:01.249135+00:00"}