{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:V6B3FAPR2ROTKKJDQT53BTVMU3","short_pith_number":"pith:V6B3FAPR","schema_version":"1.0","canonical_sha256":"af83b281f1d45d35292384fbb0ceaca6cb754cef597159e6210ca6ae6a96bd32","source":{"kind":"arxiv","id":"2412.16633","version":3},"attestation_state":"computed","paper":{"title":"POEX: Towards Policy Executable Jailbreak Attacks Against the LLM-based Robots","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.CY"],"primary_cat":"cs.RO","authors_text":"Chi Zhang, Wenyuan Xu, Xiaoyu Ji, Xinfeng Li, Xuancun Lu, Zhengxian Huang","submitted_at":"2024-12-21T13:58:27Z","abstract_excerpt":"The integration of LLMs into robots has witnessed significant growth, where LLMs can convert instructions into executable robot policies. However, the inherent vulnerability of LLMs to jailbreak attacks brings critical security risks from the digital domain to the physical world. An attacked LLM-based robot could execute harmful policies and cause physical harm. In this paper, we investigate the feasibility and rationale of jailbreak attacks against LLM-based robots and answer three research questions: (1) How applicable are existing LLM jailbreak attacks against LLM-based robots? (2) What uni"},"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":"2412.16633","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.RO","submitted_at":"2024-12-21T13:58:27Z","cross_cats_sorted":["cs.AI","cs.CY"],"title_canon_sha256":"7d42293b71bb0c0dd80de93c786940cd3b772776ea01fc39b0c4e36bd7388270","abstract_canon_sha256":"af1cda2e279dd82ad46655d9122da16c7a03bd426f11ae14bebecb9abbc9a51a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:51:20.310766Z","signature_b64":"iI2yPezxYHljbnbHbmWD+1wCXNE6SL3m2sejvQLB5QjKEN5tfk5706pTyZKBeD0iUZFTxMMZF17dP+i0Au9IDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"af83b281f1d45d35292384fbb0ceaca6cb754cef597159e6210ca6ae6a96bd32","last_reissued_at":"2026-07-05T11:51:20.310206Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:51:20.310206Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"POEX: Towards Policy Executable Jailbreak Attacks Against the LLM-based Robots","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.CY"],"primary_cat":"cs.RO","authors_text":"Chi Zhang, Wenyuan Xu, Xiaoyu Ji, Xinfeng Li, Xuancun Lu, Zhengxian Huang","submitted_at":"2024-12-21T13:58:27Z","abstract_excerpt":"The integration of LLMs into robots has witnessed significant growth, where LLMs can convert instructions into executable robot policies. However, the inherent vulnerability of LLMs to jailbreak attacks brings critical security risks from the digital domain to the physical world. An attacked LLM-based robot could execute harmful policies and cause physical harm. In this paper, we investigate the feasibility and rationale of jailbreak attacks against LLM-based robots and answer three research questions: (1) How applicable are existing LLM jailbreak attacks against LLM-based robots? (2) What uni"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.16633","kind":"arxiv","version":3},"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/2412.16633/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":"2412.16633","created_at":"2026-07-05T11:51:20.310274+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.16633v3","created_at":"2026-07-05T11:51:20.310274+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.16633","created_at":"2026-07-05T11:51:20.310274+00:00"},{"alias_kind":"pith_short_12","alias_value":"V6B3FAPR2ROT","created_at":"2026-07-05T11:51:20.310274+00:00"},{"alias_kind":"pith_short_16","alias_value":"V6B3FAPR2ROTKKJD","created_at":"2026-07-05T11:51:20.310274+00:00"},{"alias_kind":"pith_short_8","alias_value":"V6B3FAPR","created_at":"2026-07-05T11:51:20.310274+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.10749","citing_title":"Toward Secure LLM Agents: Threat Surfaces, Attacks, Defenses, and Evaluation","ref_index":111,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20544","citing_title":"The Yes-Man Syndrome: Benchmarking Abstention in Embodied Robotic Agents","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15641","citing_title":"Propagating Unsafe Actions in LLM Controlled Multi-Robot Collaboration via Single Robot Compromise","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2603.24935","citing_title":"SABER: A Stealthy Agentic Black-Box Attack Framework for Vision-Language-Action Models","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02900","citing_title":"Safety in Embodied AI: A Survey of Risks, Attacks, and Defenses","ref_index":244,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08612","citing_title":"ATAAT: Adaptive Threat-Aware Adversarial Tuning Framework against Backdoor Attacks on Vision-Language-Action Models","ref_index":33,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/V6B3FAPR2ROTKKJDQT53BTVMU3","json":"https://pith.science/pith/V6B3FAPR2ROTKKJDQT53BTVMU3.json","graph_json":"https://pith.science/api/pith-number/V6B3FAPR2ROTKKJDQT53BTVMU3/graph.json","events_json":"https://pith.science/api/pith-number/V6B3FAPR2ROTKKJDQT53BTVMU3/events.json","paper":"https://pith.science/paper/V6B3FAPR"},"agent_actions":{"view_html":"https://pith.science/pith/V6B3FAPR2ROTKKJDQT53BTVMU3","download_json":"https://pith.science/pith/V6B3FAPR2ROTKKJDQT53BTVMU3.json","view_paper":"https://pith.science/paper/V6B3FAPR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.16633&json=true","fetch_graph":"https://pith.science/api/pith-number/V6B3FAPR2ROTKKJDQT53BTVMU3/graph.json","fetch_events":"https://pith.science/api/pith-number/V6B3FAPR2ROTKKJDQT53BTVMU3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/V6B3FAPR2ROTKKJDQT53BTVMU3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/V6B3FAPR2ROTKKJDQT53BTVMU3/action/storage_attestation","attest_author":"https://pith.science/pith/V6B3FAPR2ROTKKJDQT53BTVMU3/action/author_attestation","sign_citation":"https://pith.science/pith/V6B3FAPR2ROTKKJDQT53BTVMU3/action/citation_signature","submit_replication":"https://pith.science/pith/V6B3FAPR2ROTKKJDQT53BTVMU3/action/replication_record"}},"created_at":"2026-07-05T11:51:20.310274+00:00","updated_at":"2026-07-05T11:51:20.310274+00:00"}