{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:I2PMJS3IYJLMOXWJ6PIQ7YFYJJ","short_pith_number":"pith:I2PMJS3I","schema_version":"1.0","canonical_sha256":"469ec4cb68c256c75ec9f3d10fe0b84a570db94295868c65560b6e9394c411e1","source":{"kind":"arxiv","id":"2606.07805","version":1},"attestation_state":"computed","paper":{"title":"Beyond Goodhart's Law: A Dynamic Benchmark for Evaluating Compliance in Multi-Agent Systems","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.MA"],"primary_cat":"cs.AI","authors_text":"Lizhen Qu, Qingxuan Le, Yiyang Zhao, Zenglin Xu, Zhuo Zhang","submitted_at":"2026-06-05T19:33:58Z","abstract_excerpt":"The rapid evolution of Large Language Models (LLMs) from passive assistants to autonomous, execution-capable agents has introduced critical operational risks. Most current evaluation frameworks neglect procedural compliance, leading to ''Machiavellian'' behaviors where agents strategically violate safety rules to maximize rewards - a direct manifestation of Goodhart's Law. To address this blind spot, we introduce MAC-Bench, a dynamic, adversarial benchmark designed to evaluate the procedural alignment of multi-agent systems under realistic pressure. We propose the SERV(Seed - Evolve - Refine -"},"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":"2606.07805","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-05T19:33:58Z","cross_cats_sorted":["cs.MA"],"title_canon_sha256":"9cdc472148ed7056ebd03073eda2ac49e72646c486c1a8906e8d11fac60c0dc2","abstract_canon_sha256":"30c1182357c09872da51e4edd9a09e8a732b7e5a7e32b11cbbb938983ba2a9fa"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T01:04:52.661282Z","signature_b64":"vBp1eu31+bZZtGjFM52a4hiHmHmn52dEJUlqn6LzYFcMmMNJixBiYA/PxHFoW6b81Nm6KSoeR7vT+MxnGQGTDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"469ec4cb68c256c75ec9f3d10fe0b84a570db94295868c65560b6e9394c411e1","last_reissued_at":"2026-06-09T01:04:52.660892Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T01:04:52.660892Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond Goodhart's Law: A Dynamic Benchmark for Evaluating Compliance in Multi-Agent Systems","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.MA"],"primary_cat":"cs.AI","authors_text":"Lizhen Qu, Qingxuan Le, Yiyang Zhao, Zenglin Xu, Zhuo Zhang","submitted_at":"2026-06-05T19:33:58Z","abstract_excerpt":"The rapid evolution of Large Language Models (LLMs) from passive assistants to autonomous, execution-capable agents has introduced critical operational risks. Most current evaluation frameworks neglect procedural compliance, leading to ''Machiavellian'' behaviors where agents strategically violate safety rules to maximize rewards - a direct manifestation of Goodhart's Law. To address this blind spot, we introduce MAC-Bench, a dynamic, adversarial benchmark designed to evaluate the procedural alignment of multi-agent systems under realistic pressure. We propose the SERV(Seed - Evolve - Refine -"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.07805","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/2606.07805/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":"2606.07805","created_at":"2026-06-09T01:04:52.660945+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.07805v1","created_at":"2026-06-09T01:04:52.660945+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.07805","created_at":"2026-06-09T01:04:52.660945+00:00"},{"alias_kind":"pith_short_12","alias_value":"I2PMJS3IYJLM","created_at":"2026-06-09T01:04:52.660945+00:00"},{"alias_kind":"pith_short_16","alias_value":"I2PMJS3IYJLMOXWJ","created_at":"2026-06-09T01:04:52.660945+00:00"},{"alias_kind":"pith_short_8","alias_value":"I2PMJS3I","created_at":"2026-06-09T01:04:52.660945+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/I2PMJS3IYJLMOXWJ6PIQ7YFYJJ","json":"https://pith.science/pith/I2PMJS3IYJLMOXWJ6PIQ7YFYJJ.json","graph_json":"https://pith.science/api/pith-number/I2PMJS3IYJLMOXWJ6PIQ7YFYJJ/graph.json","events_json":"https://pith.science/api/pith-number/I2PMJS3IYJLMOXWJ6PIQ7YFYJJ/events.json","paper":"https://pith.science/paper/I2PMJS3I"},"agent_actions":{"view_html":"https://pith.science/pith/I2PMJS3IYJLMOXWJ6PIQ7YFYJJ","download_json":"https://pith.science/pith/I2PMJS3IYJLMOXWJ6PIQ7YFYJJ.json","view_paper":"https://pith.science/paper/I2PMJS3I","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.07805&json=true","fetch_graph":"https://pith.science/api/pith-number/I2PMJS3IYJLMOXWJ6PIQ7YFYJJ/graph.json","fetch_events":"https://pith.science/api/pith-number/I2PMJS3IYJLMOXWJ6PIQ7YFYJJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/I2PMJS3IYJLMOXWJ6PIQ7YFYJJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/I2PMJS3IYJLMOXWJ6PIQ7YFYJJ/action/storage_attestation","attest_author":"https://pith.science/pith/I2PMJS3IYJLMOXWJ6PIQ7YFYJJ/action/author_attestation","sign_citation":"https://pith.science/pith/I2PMJS3IYJLMOXWJ6PIQ7YFYJJ/action/citation_signature","submit_replication":"https://pith.science/pith/I2PMJS3IYJLMOXWJ6PIQ7YFYJJ/action/replication_record"}},"created_at":"2026-06-09T01:04:52.660945+00:00","updated_at":"2026-06-09T01:04:52.660945+00:00"}