{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:WFY4KFICAIH4ORACID3QQRBUDC","short_pith_number":"pith:WFY4KFIC","schema_version":"1.0","canonical_sha256":"b171c51502020fc7440240f7084434189170d0e7aa28ada3ae83041447392d9c","source":{"kind":"arxiv","id":"2505.09698","version":2},"attestation_state":"computed","paper":{"title":"ManipBench: Benchmarking Vision-Language Models for Low-Level Robot Manipulation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Daniel Seita, Enyu Zhao, Hejia Zhang, Jiageng Mao, Stefanos Nikolaidis, Vedant Raval, Yue Wang, Zeyu Shangguan","submitted_at":"2025-05-14T18:01:00Z","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"},"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":"2505.09698","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2025-05-14T18:01:00Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"321acdbb2b032192f4686b6ef0cef69cbd4866b6f954ce41276deb8ea1613609","abstract_canon_sha256":"082b4f457f0f4cb7ba758f505c39c9856f136ab131636271a31d01ad8216064a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:02:01.450026Z","signature_b64":"16PL7BKGkLGbiEkZ6QOqfM3SqngfHwnSbVyQ51UBC6J6VLUV24f0A3/Lqjh+0whYs1/ZIN7z7sCp0q2RGpRtDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b171c51502020fc7440240f7084434189170d0e7aa28ada3ae83041447392d9c","last_reissued_at":"2026-07-05T12:02:01.449514Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:02:01.449514Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ManipBench: Benchmarking Vision-Language Models for Low-Level Robot Manipulation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Daniel Seita, Enyu Zhao, Hejia Zhang, Jiageng Mao, Stefanos Nikolaidis, Vedant Raval, Yue Wang, Zeyu Shangguan","submitted_at":"2025-05-14T18:01:00Z","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"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.09698","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/2505.09698/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":"2505.09698","created_at":"2026-07-05T12:02:01.449578+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.09698v2","created_at":"2026-07-05T12:02:01.449578+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.09698","created_at":"2026-07-05T12:02:01.449578+00:00"},{"alias_kind":"pith_short_12","alias_value":"WFY4KFICAIH4","created_at":"2026-07-05T12:02:01.449578+00:00"},{"alias_kind":"pith_short_16","alias_value":"WFY4KFICAIH4ORAC","created_at":"2026-07-05T12:02:01.449578+00:00"},{"alias_kind":"pith_short_8","alias_value":"WFY4KFIC","created_at":"2026-07-05T12:02:01.449578+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.08542","citing_title":"When Video Misreads: Closed-Loop Distillation of Reading Heuristics for Exploratory Manipulation Trace QA","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2605.27759","citing_title":"Colosseum V2: Benchmarking Generalization for Vision Language Action Models","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15951","citing_title":"From Failure to Feedback: Group Revision Unlocks Hard Cases in Object-Level Grounding","ref_index":100,"is_internal_anchor":false},{"citing_arxiv_id":"2511.16857","citing_title":"BOP-ASK: Object-Interaction Reasoning for Vision-Language Models","ref_index":64,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WFY4KFICAIH4ORACID3QQRBUDC","json":"https://pith.science/pith/WFY4KFICAIH4ORACID3QQRBUDC.json","graph_json":"https://pith.science/api/pith-number/WFY4KFICAIH4ORACID3QQRBUDC/graph.json","events_json":"https://pith.science/api/pith-number/WFY4KFICAIH4ORACID3QQRBUDC/events.json","paper":"https://pith.science/paper/WFY4KFIC"},"agent_actions":{"view_html":"https://pith.science/pith/WFY4KFICAIH4ORACID3QQRBUDC","download_json":"https://pith.science/pith/WFY4KFICAIH4ORACID3QQRBUDC.json","view_paper":"https://pith.science/paper/WFY4KFIC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.09698&json=true","fetch_graph":"https://pith.science/api/pith-number/WFY4KFICAIH4ORACID3QQRBUDC/graph.json","fetch_events":"https://pith.science/api/pith-number/WFY4KFICAIH4ORACID3QQRBUDC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WFY4KFICAIH4ORACID3QQRBUDC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WFY4KFICAIH4ORACID3QQRBUDC/action/storage_attestation","attest_author":"https://pith.science/pith/WFY4KFICAIH4ORACID3QQRBUDC/action/author_attestation","sign_citation":"https://pith.science/pith/WFY4KFICAIH4ORACID3QQRBUDC/action/citation_signature","submit_replication":"https://pith.science/pith/WFY4KFICAIH4ORACID3QQRBUDC/action/replication_record"}},"created_at":"2026-07-05T12:02:01.449578+00:00","updated_at":"2026-07-05T12:02:01.449578+00:00"}