{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:TJU3DQBJPPIMQBPSBQZ2T5C7SG","short_pith_number":"pith:TJU3DQBJ","schema_version":"1.0","canonical_sha256":"9a69b1c0297bd0c805f20c33a9f45f91ae742b1664adabe2edc1d8390fe9f46a","source":{"kind":"arxiv","id":"2503.19355","version":2},"attestation_state":"computed","paper":{"title":"ST-VLM: Kinematic Instruction Tuning for Spatio-Temporal Reasoning in Vision-Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dohwan Ko, Hyunwoo J. Kim, Manmohan Chandraker, Minseo Yoon, Sihyeon Kim, Vijay Kumar B.G, Yumin Suh","submitted_at":"2025-03-25T05:08:06Z","abstract_excerpt":"Spatio-temporal reasoning is essential in understanding real-world environments in various fields, eg, autonomous driving and sports analytics. Recent advances have improved the spatial reasoning ability of Vision-Language Models (VLMs) by introducing large-scale data, but these models still struggle to analyze kinematic elements like traveled distance and speed of moving objects. To bridge this gap, we construct a spatio-temporal reasoning dataset and benchmark involving kinematic instruction tuning, referred to as STKit and STKit-Bench. They consist of real-world videos with 3D annotations, "},"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":"2503.19355","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-25T05:08:06Z","cross_cats_sorted":[],"title_canon_sha256":"f98063d6a8bfeb4b55f21e68a65b1ac2f3bde2e8fed851f36ee5b7c4a4f61cd6","abstract_canon_sha256":"521c3aba7e475fce78c108a53dd750101227b81fbcc2e44736ed2fd1352be748"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:39:28.478031Z","signature_b64":"kBOg9LpnIVjrlJW91dmT/WJgFtt6R+3jQSNfT1xlZ+83jHq7VjHgh2VGruZXtnv3jxXPdD5WsOhzUL0HFpvsAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9a69b1c0297bd0c805f20c33a9f45f91ae742b1664adabe2edc1d8390fe9f46a","last_reissued_at":"2026-07-05T10:39:28.477569Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:39:28.477569Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ST-VLM: Kinematic Instruction Tuning for Spatio-Temporal Reasoning in Vision-Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dohwan Ko, Hyunwoo J. Kim, Manmohan Chandraker, Minseo Yoon, Sihyeon Kim, Vijay Kumar B.G, Yumin Suh","submitted_at":"2025-03-25T05:08:06Z","abstract_excerpt":"Spatio-temporal reasoning is essential in understanding real-world environments in various fields, eg, autonomous driving and sports analytics. Recent advances have improved the spatial reasoning ability of Vision-Language Models (VLMs) by introducing large-scale data, but these models still struggle to analyze kinematic elements like traveled distance and speed of moving objects. To bridge this gap, we construct a spatio-temporal reasoning dataset and benchmark involving kinematic instruction tuning, referred to as STKit and STKit-Bench. They consist of real-world videos with 3D annotations, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.19355","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/2503.19355/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":"2503.19355","created_at":"2026-07-05T10:39:28.477622+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.19355v2","created_at":"2026-07-05T10:39:28.477622+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.19355","created_at":"2026-07-05T10:39:28.477622+00:00"},{"alias_kind":"pith_short_12","alias_value":"TJU3DQBJPPIM","created_at":"2026-07-05T10:39:28.477622+00:00"},{"alias_kind":"pith_short_16","alias_value":"TJU3DQBJPPIMQBPS","created_at":"2026-07-05T10:39:28.477622+00:00"},{"alias_kind":"pith_short_8","alias_value":"TJU3DQBJ","created_at":"2026-07-05T10:39:28.477622+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.06750","citing_title":"How Well Do Vision-Language Models Understand Sequential Driving Scenes? A Sensitivity Study","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19382","citing_title":"PRISM: A Benchmark for Programmatic Spatial-Temporal Reasoning","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2512.17012","citing_title":"4D-RGPT: Toward Region-level 4D Understanding via Perceptual Distillation","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06750","citing_title":"How Well Do Vision-Language Models Understand Sequential Driving Scenes? A Sensitivity Study","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TJU3DQBJPPIMQBPSBQZ2T5C7SG","json":"https://pith.science/pith/TJU3DQBJPPIMQBPSBQZ2T5C7SG.json","graph_json":"https://pith.science/api/pith-number/TJU3DQBJPPIMQBPSBQZ2T5C7SG/graph.json","events_json":"https://pith.science/api/pith-number/TJU3DQBJPPIMQBPSBQZ2T5C7SG/events.json","paper":"https://pith.science/paper/TJU3DQBJ"},"agent_actions":{"view_html":"https://pith.science/pith/TJU3DQBJPPIMQBPSBQZ2T5C7SG","download_json":"https://pith.science/pith/TJU3DQBJPPIMQBPSBQZ2T5C7SG.json","view_paper":"https://pith.science/paper/TJU3DQBJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.19355&json=true","fetch_graph":"https://pith.science/api/pith-number/TJU3DQBJPPIMQBPSBQZ2T5C7SG/graph.json","fetch_events":"https://pith.science/api/pith-number/TJU3DQBJPPIMQBPSBQZ2T5C7SG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TJU3DQBJPPIMQBPSBQZ2T5C7SG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TJU3DQBJPPIMQBPSBQZ2T5C7SG/action/storage_attestation","attest_author":"https://pith.science/pith/TJU3DQBJPPIMQBPSBQZ2T5C7SG/action/author_attestation","sign_citation":"https://pith.science/pith/TJU3DQBJPPIMQBPSBQZ2T5C7SG/action/citation_signature","submit_replication":"https://pith.science/pith/TJU3DQBJPPIMQBPSBQZ2T5C7SG/action/replication_record"}},"created_at":"2026-07-05T10:39:28.477622+00:00","updated_at":"2026-07-05T10:39:28.477622+00:00"}