{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:SUSBJD7FTA3HL6AV7TUXN6EAFW","short_pith_number":"pith:SUSBJD7F","schema_version":"1.0","canonical_sha256":"9524148fe5983675f815fce976f8802d944446a5fba41e652d8a99874573fe0b","source":{"kind":"arxiv","id":"2504.13399","version":1},"attestation_state":"computed","paper":{"title":"Towards a Multi-Agent Vision-Language System for Zero-Shot Novel Hazardous Object Detection for Autonomous Driving Safety","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Andreas M{\\o}gelmose, Aryan Keskar, Harsha Krishnaswamy, Ross Greer, Shashank Shriram, Srinivasa Perisetla, Tonko Emil Westerhof Bossen","submitted_at":"2025-04-18T01:25:02Z","abstract_excerpt":"Detecting anomalous hazards in visual data, particularly in video streams, is a critical challenge in autonomous driving. Existing models often struggle with unpredictable, out-of-label hazards due to their reliance on predefined object categories. In this paper, we propose a multimodal approach that integrates vision-language reasoning with zero-shot object detection to improve hazard identification and explanation. Our pipeline consists of a Vision-Language Model (VLM), a Large Language Model (LLM), in order to detect hazardous objects within a traffic scene. We refine object detection by in"},"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":"2504.13399","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-04-18T01:25:02Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ec3a67b22e052129ee8220e8f85bdf1c61d4930259d432c98b0080ec63599ab8","abstract_canon_sha256":"42ba0183f87894d23fb07cd3676e631cb32c132bda1619f36be621f1ad914614"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:50:54.130936Z","signature_b64":"bLASz00K7rsJDbWcCpt1M9r8kmm5RjUL2zKPHn/UVO7j497ptfUU5djcJ/ibiH22NTeUMg2DQp5F5QGH+FBIAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9524148fe5983675f815fce976f8802d944446a5fba41e652d8a99874573fe0b","last_reissued_at":"2026-07-05T10:50:54.130448Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:50:54.130448Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards a Multi-Agent Vision-Language System for Zero-Shot Novel Hazardous Object Detection for Autonomous Driving Safety","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Andreas M{\\o}gelmose, Aryan Keskar, Harsha Krishnaswamy, Ross Greer, Shashank Shriram, Srinivasa Perisetla, Tonko Emil Westerhof Bossen","submitted_at":"2025-04-18T01:25:02Z","abstract_excerpt":"Detecting anomalous hazards in visual data, particularly in video streams, is a critical challenge in autonomous driving. Existing models often struggle with unpredictable, out-of-label hazards due to their reliance on predefined object categories. In this paper, we propose a multimodal approach that integrates vision-language reasoning with zero-shot object detection to improve hazard identification and explanation. Our pipeline consists of a Vision-Language Model (VLM), a Large Language Model (LLM), in order to detect hazardous objects within a traffic scene. We refine object detection by in"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.13399","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/2504.13399/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":"2504.13399","created_at":"2026-07-05T10:50:54.130507+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.13399v1","created_at":"2026-07-05T10:50:54.130507+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.13399","created_at":"2026-07-05T10:50:54.130507+00:00"},{"alias_kind":"pith_short_12","alias_value":"SUSBJD7FTA3H","created_at":"2026-07-05T10:50:54.130507+00:00"},{"alias_kind":"pith_short_16","alias_value":"SUSBJD7FTA3HL6AV","created_at":"2026-07-05T10:50:54.130507+00:00"},{"alias_kind":"pith_short_8","alias_value":"SUSBJD7F","created_at":"2026-07-05T10:50:54.130507+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.03631","citing_title":"Single-agent vs. Multi-agents for Automated Video Analysis of On-Screen Collaborative Learning Behaviors","ref_index":24,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SUSBJD7FTA3HL6AV7TUXN6EAFW","json":"https://pith.science/pith/SUSBJD7FTA3HL6AV7TUXN6EAFW.json","graph_json":"https://pith.science/api/pith-number/SUSBJD7FTA3HL6AV7TUXN6EAFW/graph.json","events_json":"https://pith.science/api/pith-number/SUSBJD7FTA3HL6AV7TUXN6EAFW/events.json","paper":"https://pith.science/paper/SUSBJD7F"},"agent_actions":{"view_html":"https://pith.science/pith/SUSBJD7FTA3HL6AV7TUXN6EAFW","download_json":"https://pith.science/pith/SUSBJD7FTA3HL6AV7TUXN6EAFW.json","view_paper":"https://pith.science/paper/SUSBJD7F","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.13399&json=true","fetch_graph":"https://pith.science/api/pith-number/SUSBJD7FTA3HL6AV7TUXN6EAFW/graph.json","fetch_events":"https://pith.science/api/pith-number/SUSBJD7FTA3HL6AV7TUXN6EAFW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SUSBJD7FTA3HL6AV7TUXN6EAFW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SUSBJD7FTA3HL6AV7TUXN6EAFW/action/storage_attestation","attest_author":"https://pith.science/pith/SUSBJD7FTA3HL6AV7TUXN6EAFW/action/author_attestation","sign_citation":"https://pith.science/pith/SUSBJD7FTA3HL6AV7TUXN6EAFW/action/citation_signature","submit_replication":"https://pith.science/pith/SUSBJD7FTA3HL6AV7TUXN6EAFW/action/replication_record"}},"created_at":"2026-07-05T10:50:54.130507+00:00","updated_at":"2026-07-05T10:50:54.130507+00:00"}