{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:WFY4KFICAIH4ORACID3QQRBUDC","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"082b4f457f0f4cb7ba758f505c39c9856f136ab131636271a31d01ad8216064a","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2025-05-14T18:01:00Z","title_canon_sha256":"321acdbb2b032192f4686b6ef0cef69cbd4866b6f954ce41276deb8ea1613609"},"schema_version":"1.0","source":{"id":"2505.09698","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.09698","created_at":"2026-07-05T12:02:01Z"},{"alias_kind":"arxiv_version","alias_value":"2505.09698v2","created_at":"2026-07-05T12:02:01Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.09698","created_at":"2026-07-05T12:02:01Z"},{"alias_kind":"pith_short_12","alias_value":"WFY4KFICAIH4","created_at":"2026-07-05T12:02:01Z"},{"alias_kind":"pith_short_16","alias_value":"WFY4KFICAIH4ORAC","created_at":"2026-07-05T12:02:01Z"},{"alias_kind":"pith_short_8","alias_value":"WFY4KFIC","created_at":"2026-07-05T12:02:01Z"}],"graph_snapshots":[{"event_id":"sha256:b982fc06522793414e5a62269bf0f499ab437ae7a0d6389aea26f304d72be628","target":"graph","created_at":"2026-07-05T12:02:01Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2505.09698/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Vision-Language Models (VLMs) have revolutionized artificial intelligence and robotics due to their commonsense reasoning capabilities. In robotic manipulation, VLMs are used primarily as high-level planners, but recent work has also studied their lower-level reasoning ability, which refers to making decisions about precise robot movements. However, the community currently lacks a clear and common benchmark that can evaluate how well VLMs can aid low-level reasoning in robotics. Consequently, we propose a novel benchmark, ManipBench, to evaluate the low-level robot manipulation reasoning capab","authors_text":"Daniel Seita, Enyu Zhao, Hejia Zhang, Jiageng Mao, Stefanos Nikolaidis, Vedant Raval, Yue Wang, Zeyu Shangguan","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2025-05-14T18:01:00Z","title":"ManipBench: Benchmarking Vision-Language Models for Low-Level Robot Manipulation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.09698","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:f49f7203b2be494e945bb7a6ede203c7cdf7dc4ce26555910715c957cd619381","target":"record","created_at":"2026-07-05T12:02:01Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"082b4f457f0f4cb7ba758f505c39c9856f136ab131636271a31d01ad8216064a","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2025-05-14T18:01:00Z","title_canon_sha256":"321acdbb2b032192f4686b6ef0cef69cbd4866b6f954ce41276deb8ea1613609"},"schema_version":"1.0","source":{"id":"2505.09698","kind":"arxiv","version":2}},"canonical_sha256":"b171c51502020fc7440240f7084434189170d0e7aa28ada3ae83041447392d9c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b171c51502020fc7440240f7084434189170d0e7aa28ada3ae83041447392d9c","first_computed_at":"2026-07-05T12:02:01.449514Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:02:01.449514Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"16PL7BKGkLGbiEkZ6QOqfM3SqngfHwnSbVyQ51UBC6J6VLUV24f0A3/Lqjh+0whYs1/ZIN7z7sCp0q2RGpRtDA==","signature_status":"signed_v1","signed_at":"2026-07-05T12:02:01.450026Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.09698","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f49f7203b2be494e945bb7a6ede203c7cdf7dc4ce26555910715c957cd619381","sha256:b982fc06522793414e5a62269bf0f499ab437ae7a0d6389aea26f304d72be628"],"state_sha256":"09ccc0c923960d7a253e258e91a98fa85429981eb4d31c69d77c082cbbc0ef79"}