{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:WEQNK2C35XVXGASYJ6D4254MUR","short_pith_number":"pith:WEQNK2C3","schema_version":"1.0","canonical_sha256":"b120d5685bedeb7302584f87cd778ca4713e424b1a0b83106edd11cad2b4effa","source":{"kind":"arxiv","id":"2401.10019","version":3},"attestation_state":"computed","paper":{"title":"R-Judge: Benchmarking Safety Risk Awareness for LLM Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Binglin Zhou, Fangqi Li, Gongshen Liu, Lingzhong Dong, Lizhen Xu, Ruijie Zhao, Rui Wang, Tian Xia, Tongxin Yuan, Yiming Wang, Zhiwei He, Zhuosheng Zhang","submitted_at":"2024-01-18T14:40:46Z","abstract_excerpt":"Large language models (LLMs) have exhibited great potential in autonomously completing tasks across real-world applications. Despite this, these LLM agents introduce unexpected safety risks when operating in interactive environments. Instead of centering on the harmlessness of LLM-generated content in most prior studies, this work addresses the imperative need for benchmarking the behavioral safety of LLM agents within diverse environments. We introduce R-Judge, a benchmark crafted to evaluate the proficiency of LLMs in judging and identifying safety risks given agent interaction records. R-Ju"},"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":"2401.10019","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-01-18T14:40:46Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f74abca7b0e8ad251fb1f64116e6184d6f3fc93c0989ecb0278fa63a4dd7a5bd","abstract_canon_sha256":"b0072f11e20f99b1848b2f8090070539128f82f1c5a5b7e6ffc175fdc82dec0d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:16:11.737968Z","signature_b64":"p9P0mtA90rfh8tlJIMowIVZP+7zSqET0rqn7Ee63HIosuO8dX4yYMiZafYa35jXVeiMCMMvNKupIE04JlhI7Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b120d5685bedeb7302584f87cd778ca4713e424b1a0b83106edd11cad2b4effa","last_reissued_at":"2026-07-05T09:16:11.737407Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:16:11.737407Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"R-Judge: Benchmarking Safety Risk Awareness for LLM Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Binglin Zhou, Fangqi Li, Gongshen Liu, Lingzhong Dong, Lizhen Xu, Ruijie Zhao, Rui Wang, Tian Xia, Tongxin Yuan, Yiming Wang, Zhiwei He, Zhuosheng Zhang","submitted_at":"2024-01-18T14:40:46Z","abstract_excerpt":"Large language models (LLMs) have exhibited great potential in autonomously completing tasks across real-world applications. Despite this, these LLM agents introduce unexpected safety risks when operating in interactive environments. Instead of centering on the harmlessness of LLM-generated content in most prior studies, this work addresses the imperative need for benchmarking the behavioral safety of LLM agents within diverse environments. We introduce R-Judge, a benchmark crafted to evaluate the proficiency of LLMs in judging and identifying safety risks given agent interaction records. R-Ju"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.10019","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/2401.10019/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":"2401.10019","created_at":"2026-07-05T09:16:11.737473+00:00"},{"alias_kind":"arxiv_version","alias_value":"2401.10019v3","created_at":"2026-07-05T09:16:11.737473+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.10019","created_at":"2026-07-05T09:16:11.737473+00:00"},{"alias_kind":"pith_short_12","alias_value":"WEQNK2C35XVX","created_at":"2026-07-05T09:16:11.737473+00:00"},{"alias_kind":"pith_short_16","alias_value":"WEQNK2C35XVXGASY","created_at":"2026-07-05T09:16:11.737473+00:00"},{"alias_kind":"pith_short_8","alias_value":"WEQNK2C3","created_at":"2026-07-05T09:16:11.737473+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":19,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2607.05743","citing_title":"The Balkanization of Execution-Security Research for AI Coding Agents: Isolation, Access Control, and Time-of-Check-to-Time-of-Use Vulnerabilities","ref_index":50,"is_internal_anchor":true},{"citing_arxiv_id":"2607.07676","citing_title":"SkillCenter: A Large-Scale Source-Grounded Skill Library for Autonomous AI Agents","ref_index":35,"is_internal_anchor":true},{"citing_arxiv_id":"2606.19887","citing_title":"FinRED: An Expert-Guided Benchmark Generation and Evaluation Framework for Financial LLM Red-Teaming","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2606.18550","citing_title":"The Gate Is Only as Honest as Its Contracts: ContractGuard for the Contract Layer of Risk-Aware Causal Gating","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2606.19380","citing_title":"ClayBuddy: A Framework, Evaluation, & Mitigation of Coding Agent Failures","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2606.10484","citing_title":"AgentCanary: A Security Evaluation Framework for Autonomous AI Agents in Real Executable Environments","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09255","citing_title":"RPO-PDT: Demonstrating Role-Play-Based Knowledge Adaptation for Student Support Dialogue (Demonstration System)","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2606.04037","citing_title":"Toward Pre-Deployment Assurance for Enterprise AI Agents: Ontology-Grounded Simulation and Trust Certification","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2606.02668","citing_title":"What You Approve Is What Executes: Consent Integrity for Black-Box LLM Agents","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16282","citing_title":"Taxonomy and Consistency Analysis of Safety Benchmarks for AI Agents","ref_index":62,"is_internal_anchor":false},{"citing_arxiv_id":"2402.03578","citing_title":"LLM Multi-Agent Systems: Challenges and Open Problems","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2604.03242","citing_title":"DRAFT: Task Decoupled Latent Reasoning for Agent Safety","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2504.19678","citing_title":"From LLM Reasoning to Autonomous AI Agents: A Comprehensive Review","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02022","citing_title":"ATBench: A Diverse and Realistic Agent Trajectory Benchmark for Safety Evaluation and Diagnosis","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02022","citing_title":"ATBench: A Diverse and Realistic Agent Trajectory Benchmark for Safety Evaluation and Diagnosis","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2410.02644","citing_title":"Agent Security Bench (ASB): Formalizing and Benchmarking Attacks and Defenses in LLM-based Agents","ref_index":63,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10223","citing_title":"Beyond Autonomy: A Dynamic Tiered AgentRunner Framework for Governable and Resilient Enterprise AI Execution","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12177","citing_title":"Policy-Invisible Violations in LLM-Based Agents","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2604.09056","citing_title":"Conversations Risk Detection LLMs in Financial Agents via Multi-Stage Generative Rollout","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WEQNK2C35XVXGASYJ6D4254MUR","json":"https://pith.science/pith/WEQNK2C35XVXGASYJ6D4254MUR.json","graph_json":"https://pith.science/api/pith-number/WEQNK2C35XVXGASYJ6D4254MUR/graph.json","events_json":"https://pith.science/api/pith-number/WEQNK2C35XVXGASYJ6D4254MUR/events.json","paper":"https://pith.science/paper/WEQNK2C3"},"agent_actions":{"view_html":"https://pith.science/pith/WEQNK2C35XVXGASYJ6D4254MUR","download_json":"https://pith.science/pith/WEQNK2C35XVXGASYJ6D4254MUR.json","view_paper":"https://pith.science/paper/WEQNK2C3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2401.10019&json=true","fetch_graph":"https://pith.science/api/pith-number/WEQNK2C35XVXGASYJ6D4254MUR/graph.json","fetch_events":"https://pith.science/api/pith-number/WEQNK2C35XVXGASYJ6D4254MUR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WEQNK2C35XVXGASYJ6D4254MUR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WEQNK2C35XVXGASYJ6D4254MUR/action/storage_attestation","attest_author":"https://pith.science/pith/WEQNK2C35XVXGASYJ6D4254MUR/action/author_attestation","sign_citation":"https://pith.science/pith/WEQNK2C35XVXGASYJ6D4254MUR/action/citation_signature","submit_replication":"https://pith.science/pith/WEQNK2C35XVXGASYJ6D4254MUR/action/replication_record"}},"created_at":"2026-07-05T09:16:11.737473+00:00","updated_at":"2026-07-05T09:16:11.737473+00:00"}