{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:H7JIDRSOI7HAJVZZGOC3HNJDY7","short_pith_number":"pith:H7JIDRSO","schema_version":"1.0","canonical_sha256":"3fd281c64e47ce04d7393385b3b523c7c14f086e38c817febb233c095d4d26e9","source":{"kind":"arxiv","id":"2503.15707","version":1},"attestation_state":"computed","paper":{"title":"Safety Aware Task Planning via Large Language Models in Robotics","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Ali Anwar, Azal Ahmad Khan, Jie Ding, Michael Andrev, Muhammad Ali Murtaza, Rui Zhang, Sergio Aguilera, Seth Hutchinson","submitted_at":"2025-03-19T21:41:10Z","abstract_excerpt":"The integration of large language models (LLMs) into robotic task planning has unlocked better reasoning capabilities for complex, long-horizon workflows. However, ensuring safety in LLM-driven plans remains a critical challenge, as these models often prioritize task completion over risk mitigation. This paper introduces SAFER (Safety-Aware Framework for Execution in Robotics), a multi-LLM framework designed to embed safety awareness into robotic task planning. SAFER employs a Safety Agent that operates alongside the primary task planner, providing safety feedback. Additionally, we introduce L"},"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":"2503.15707","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2025-03-19T21:41:10Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b90b71190b0315f473ee4d35d70878801e498657799b0ab8aa2dd6ea3fa259cf","abstract_canon_sha256":"19a9794c4cd03713bb174c8f1e70123f68779e39b7193bf2b7387bd8b0ee6241"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:35:56.734475Z","signature_b64":"dkeRbaZMKyrD6PwYxsoSflgROdX6Mode3XubX5lMPTHYk4r5yJ/XY1+hr92ciYAZ1b2V3MiezGxL0rjEDzEtCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3fd281c64e47ce04d7393385b3b523c7c14f086e38c817febb233c095d4d26e9","last_reissued_at":"2026-07-05T10:35:56.733872Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:35:56.733872Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Safety Aware Task Planning via Large Language Models in Robotics","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Ali Anwar, Azal Ahmad Khan, Jie Ding, Michael Andrev, Muhammad Ali Murtaza, Rui Zhang, Sergio Aguilera, Seth Hutchinson","submitted_at":"2025-03-19T21:41:10Z","abstract_excerpt":"The integration of large language models (LLMs) into robotic task planning has unlocked better reasoning capabilities for complex, long-horizon workflows. However, ensuring safety in LLM-driven plans remains a critical challenge, as these models often prioritize task completion over risk mitigation. This paper introduces SAFER (Safety-Aware Framework for Execution in Robotics), a multi-LLM framework designed to embed safety awareness into robotic task planning. SAFER employs a Safety Agent that operates alongside the primary task planner, providing safety feedback. Additionally, we introduce L"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.15707","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/2503.15707/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":"2503.15707","created_at":"2026-07-05T10:35:56.733936+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.15707v1","created_at":"2026-07-05T10:35:56.733936+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.15707","created_at":"2026-07-05T10:35:56.733936+00:00"},{"alias_kind":"pith_short_12","alias_value":"H7JIDRSOI7HA","created_at":"2026-07-05T10:35:56.733936+00:00"},{"alias_kind":"pith_short_16","alias_value":"H7JIDRSOI7HAJVZZ","created_at":"2026-07-05T10:35:56.733936+00:00"},{"alias_kind":"pith_short_8","alias_value":"H7JIDRSO","created_at":"2026-07-05T10:35:56.733936+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.09416","citing_title":"Harness Engineering for Physical AI: Robot Middleware Is the Harness Layer","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18672","citing_title":"Position: A Three-Layer Probabilistic Assume-Guarantee Architecture Is Structurally Required for Safe LLM Agent Deployment","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2511.12431","citing_title":"Online Adaptive Probabilistic Safety Certificate with Language Guidance","ref_index":9,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/H7JIDRSOI7HAJVZZGOC3HNJDY7","json":"https://pith.science/pith/H7JIDRSOI7HAJVZZGOC3HNJDY7.json","graph_json":"https://pith.science/api/pith-number/H7JIDRSOI7HAJVZZGOC3HNJDY7/graph.json","events_json":"https://pith.science/api/pith-number/H7JIDRSOI7HAJVZZGOC3HNJDY7/events.json","paper":"https://pith.science/paper/H7JIDRSO"},"agent_actions":{"view_html":"https://pith.science/pith/H7JIDRSOI7HAJVZZGOC3HNJDY7","download_json":"https://pith.science/pith/H7JIDRSOI7HAJVZZGOC3HNJDY7.json","view_paper":"https://pith.science/paper/H7JIDRSO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.15707&json=true","fetch_graph":"https://pith.science/api/pith-number/H7JIDRSOI7HAJVZZGOC3HNJDY7/graph.json","fetch_events":"https://pith.science/api/pith-number/H7JIDRSOI7HAJVZZGOC3HNJDY7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/H7JIDRSOI7HAJVZZGOC3HNJDY7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/H7JIDRSOI7HAJVZZGOC3HNJDY7/action/storage_attestation","attest_author":"https://pith.science/pith/H7JIDRSOI7HAJVZZGOC3HNJDY7/action/author_attestation","sign_citation":"https://pith.science/pith/H7JIDRSOI7HAJVZZGOC3HNJDY7/action/citation_signature","submit_replication":"https://pith.science/pith/H7JIDRSOI7HAJVZZGOC3HNJDY7/action/replication_record"}},"created_at":"2026-07-05T10:35:56.733936+00:00","updated_at":"2026-07-05T10:35:56.733936+00:00"}