{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:JFL3EHACTBFBLAX2LDW2E3V7JC","short_pith_number":"pith:JFL3EHAC","schema_version":"1.0","canonical_sha256":"4957b21c02984a1582fa58eda26ebf48849efae5bd7522c9fb1809b8ae11f31d","source":{"kind":"arxiv","id":"2603.05754","version":2},"attestation_state":"computed","paper":{"title":"Safe-Night VLA: Seeing the Unseen via Thermal-Perceptive Vision-Language-Action Models for Safety-Critical Manipulation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Bingkun Huang, Dian Yu, Majid Khadiv, Qingchuan Zhou, Zewen Yang","submitted_at":"2026-03-05T23:26:44Z","abstract_excerpt":"Current Vision-Language-Action (VLA) models rely primarily on RGB perception, preventing them from capturing modalities such as thermal signals that are imperceptible to conventional visual sensors. Moreover, end-to-end generative policies lack explicit safety constraints, making them fragile when encountering obstacles and novel scenarios outside the training distribution. To address these limitations, we propose Safe-Night VLA, a multimodal manipulation framework that enables robots to see the unseen while enforcing rigorous safety constraints for thermal-aware manipulation in unstructured e"},"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":"2603.05754","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2026-03-05T23:26:44Z","cross_cats_sorted":[],"title_canon_sha256":"6452581ffa8e08dbf11409d44adb41f674a3f2e4a14e2efd5b71104054d1a8a6","abstract_canon_sha256":"a7bddd7df23558921bc1afd5fccc737deb277cf1f906fabbf6a6b64f162b44d5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-17T01:21:46.948380Z","signature_b64":"m5096flqDhOBYeOstrLXl6HJ2ciG6t0sQuYqPiukbjPEYp0LjwIcqWKRWqApruhcBD7xdZX71j8hVxXn3q/HDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4957b21c02984a1582fa58eda26ebf48849efae5bd7522c9fb1809b8ae11f31d","last_reissued_at":"2026-07-17T01:21:46.947365Z","signature_status":"signed_v1","first_computed_at":"2026-07-17T01:21:46.947365Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Safe-Night VLA: Seeing the Unseen via Thermal-Perceptive Vision-Language-Action Models for Safety-Critical Manipulation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Bingkun Huang, Dian Yu, Majid Khadiv, Qingchuan Zhou, Zewen Yang","submitted_at":"2026-03-05T23:26:44Z","abstract_excerpt":"Current Vision-Language-Action (VLA) models rely primarily on RGB perception, preventing them from capturing modalities such as thermal signals that are imperceptible to conventional visual sensors. Moreover, end-to-end generative policies lack explicit safety constraints, making them fragile when encountering obstacles and novel scenarios outside the training distribution. To address these limitations, we propose Safe-Night VLA, a multimodal manipulation framework that enables robots to see the unseen while enforcing rigorous safety constraints for thermal-aware manipulation in unstructured e"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.05754","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/2603.05754/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":"2603.05754","created_at":"2026-07-17T01:21:46.947813+00:00"},{"alias_kind":"arxiv_version","alias_value":"2603.05754v2","created_at":"2026-07-17T01:21:46.947813+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.05754","created_at":"2026-07-17T01:21:46.947813+00:00"},{"alias_kind":"pith_short_12","alias_value":"JFL3EHACTBFB","created_at":"2026-07-17T01:21:46.947813+00:00"},{"alias_kind":"pith_short_16","alias_value":"JFL3EHACTBFBLAX2","created_at":"2026-07-17T01:21:46.947813+00:00"},{"alias_kind":"pith_short_8","alias_value":"JFL3EHAC","created_at":"2026-07-17T01:21:46.947813+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2606.20698","citing_title":"SafeDojo: Safe Reinforcement Learning for VLA via Interactive World Model","ref_index":20,"is_internal_anchor":true},{"citing_arxiv_id":"2604.23775","citing_title":"Vision-Language-Action Safety: Threats, Challenges, Evaluations, and Mechanisms","ref_index":86,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JFL3EHACTBFBLAX2LDW2E3V7JC","json":"https://pith.science/pith/JFL3EHACTBFBLAX2LDW2E3V7JC.json","graph_json":"https://pith.science/api/pith-number/JFL3EHACTBFBLAX2LDW2E3V7JC/graph.json","events_json":"https://pith.science/api/pith-number/JFL3EHACTBFBLAX2LDW2E3V7JC/events.json","paper":"https://pith.science/paper/JFL3EHAC"},"agent_actions":{"view_html":"https://pith.science/pith/JFL3EHACTBFBLAX2LDW2E3V7JC","download_json":"https://pith.science/pith/JFL3EHACTBFBLAX2LDW2E3V7JC.json","view_paper":"https://pith.science/paper/JFL3EHAC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2603.05754&json=true","fetch_graph":"https://pith.science/api/pith-number/JFL3EHACTBFBLAX2LDW2E3V7JC/graph.json","fetch_events":"https://pith.science/api/pith-number/JFL3EHACTBFBLAX2LDW2E3V7JC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JFL3EHACTBFBLAX2LDW2E3V7JC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JFL3EHACTBFBLAX2LDW2E3V7JC/action/storage_attestation","attest_author":"https://pith.science/pith/JFL3EHACTBFBLAX2LDW2E3V7JC/action/author_attestation","sign_citation":"https://pith.science/pith/JFL3EHACTBFBLAX2LDW2E3V7JC/action/citation_signature","submit_replication":"https://pith.science/pith/JFL3EHACTBFBLAX2LDW2E3V7JC/action/replication_record"}},"created_at":"2026-07-17T01:21:46.947813+00:00","updated_at":"2026-07-17T01:21:46.947813+00:00"}