{"paper":{"title":"CalBench: Evaluating Coordination-Privacy Trade-offs in Multi-Agent LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"CalBench is a benchmark where agents with private calendars must coordinate meeting schedules without sharing data.","cross_cats":["cs.AI"],"primary_cat":"cs.MA","authors_text":"Chelsea Zou, Noah Goodman, Robert D. Hawkins, Selena She, Yiheng Yao","submitted_at":"2026-05-10T23:56:02Z","abstract_excerpt":"Personal AI assistants are beginning to act as delegates with access to calendars, inboxes, and user preferences. Calendar scheduling makes the trust problem concrete: an assistant must coordinate with other assistants while deciding what to reveal about the person it represents. We introduce CalBench, a controlled benchmark for multi-agent calendar scheduling under private information. In each task, $N$ agents manage separate private calendars and schedule a stream of $M$ incoming meetings while minimizing disruption costs. Because no agent can inspect another agent's calendar, success requir"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"CalBench is inherently decentralized: no agent has access to another agent's private calendar, yet agents must still reach mutually consistent decisions over shared meeting scheduling, enabling precise verification of task success, communication efficiency, and fairness.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the specific mechanics of private-calendar scheduling with oracle optima and semantic privacy tags form a representative and controllable proxy for general coordination-privacy trade-offs across multi-agent LLM applications.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"CalBench introduces a decentralized multi-agent benchmark for calendar scheduling that measures coordination quality against an oracle optimum and privacy leakage under private information constraints.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"CalBench is a benchmark where agents with private calendars must coordinate meeting schedules without sharing data.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"f2338722e2f3d872f9a3ede27eb8410c0f9b6869b2af6dd4e9da7d3568ef7ac3"},"source":{"id":"2605.09823","kind":"arxiv","version":2},"verdict":{"id":"db45bd4b-e345-48a3-865d-82027e5a1ff9","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-12T01:59:53.883962Z","strongest_claim":"CalBench is inherently decentralized: no agent has access to another agent's private calendar, yet agents must still reach mutually consistent decisions over shared meeting scheduling, enabling precise verification of task success, communication efficiency, and fairness.","one_line_summary":"CalBench introduces a decentralized multi-agent benchmark for calendar scheduling that measures coordination quality against an oracle optimum and privacy leakage under private information constraints.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the specific mechanics of private-calendar scheduling with oracle optima and semantic privacy tags form a representative and controllable proxy for general coordination-privacy trade-offs across multi-agent LLM applications.","pith_extraction_headline":"CalBench is a benchmark where agents with private calendars must coordinate meeting schedules without sharing data."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.09823/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-20T07:02:01.365544Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T16:35:15.579659Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T12:31:17.637007Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T09:54:16.808891Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"6be5f186f5a1607108e81c467ff622b4e4a25bb8d9bc38cd22b9ff1c78649415"},"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"}