{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:HFETZMBQYJ4XW3ASQPFUAJE77A","short_pith_number":"pith:HFETZMBQ","schema_version":"1.0","canonical_sha256":"39493cb030c2797b6c1283cb40249ff80c428d5c5cf4db3e0df6afb5f8c4cfb2","source":{"kind":"arxiv","id":"2608.05263","version":1},"attestation_state":"computed","paper":{"title":"OrchestraBench: Evaluating Multi-Agent Orchestration Failure Modes, Recovery, and Decomposition Quality","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Natan Vidra, Sharon Zheng, Spurthi Setty, Yidian Chen, Yingzi Gu","submitted_at":"2026-08-05T17:27:15Z","abstract_excerpt":"Multi-agent orchestration frameworks are moving from demos to production, yet benchmarks typically report task accuracy without diagnosing why a pipeline failed, where a cascade began, or which routing decision caused the breakdown. OrchestraBench evaluates failure, recovery, and decomposition through a controlled, seed-reproducible failure-injection harness over templated enterprise workflows. It introduces cascade radius and per-failure-mode recovery as primary metrics and compares routing policies with bootstrap confidence intervals and paired tests. On a 26-case gold-labelled diagnostic, a"},"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":"2608.05263","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-08-05T17:27:15Z","cross_cats_sorted":[],"title_canon_sha256":"6adaddc57b3f66a113891a96caaa837e2870d044f915ba4f005bed9724a6be25","abstract_canon_sha256":"3eb39a519db37561290da42e11fd3ebe0eaecc53fefdd912a0e14752e21f18f2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-07T00:46:38.172062Z","signature_b64":"77+110+tBcKGCC0Pi55TPzw+3NDgPSkERHwNeCBDNFsDGqoaX6WXoikDWBK9IyiV53I/Crfaxs4R4DRvTh1EAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"39493cb030c2797b6c1283cb40249ff80c428d5c5cf4db3e0df6afb5f8c4cfb2","last_reissued_at":"2026-08-07T00:46:38.170678Z","signature_status":"signed_v1","first_computed_at":"2026-08-07T00:46:38.170678Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"OrchestraBench: Evaluating Multi-Agent Orchestration Failure Modes, Recovery, and Decomposition Quality","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Natan Vidra, Sharon Zheng, Spurthi Setty, Yidian Chen, Yingzi Gu","submitted_at":"2026-08-05T17:27:15Z","abstract_excerpt":"Multi-agent orchestration frameworks are moving from demos to production, yet benchmarks typically report task accuracy without diagnosing why a pipeline failed, where a cascade began, or which routing decision caused the breakdown. OrchestraBench evaluates failure, recovery, and decomposition through a controlled, seed-reproducible failure-injection harness over templated enterprise workflows. It introduces cascade radius and per-failure-mode recovery as primary metrics and compares routing policies with bootstrap confidence intervals and paired tests. On a 26-case gold-labelled diagnostic, a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.05263","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/2608.05263/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":"2608.05263","created_at":"2026-08-07T00:46:38.171974+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.05263v1","created_at":"2026-08-07T00:46:38.171974+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.05263","created_at":"2026-08-07T00:46:38.171974+00:00"},{"alias_kind":"pith_short_12","alias_value":"HFETZMBQYJ4X","created_at":"2026-08-07T00:46:38.171974+00:00"},{"alias_kind":"pith_short_16","alias_value":"HFETZMBQYJ4XW3AS","created_at":"2026-08-07T00:46:38.171974+00:00"},{"alias_kind":"pith_short_8","alias_value":"HFETZMBQ","created_at":"2026-08-07T00:46:38.171974+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/HFETZMBQYJ4XW3ASQPFUAJE77A","json":"https://pith.science/pith/HFETZMBQYJ4XW3ASQPFUAJE77A.json","graph_json":"https://pith.science/api/pith-number/HFETZMBQYJ4XW3ASQPFUAJE77A/graph.json","events_json":"https://pith.science/api/pith-number/HFETZMBQYJ4XW3ASQPFUAJE77A/events.json","paper":"https://pith.science/paper/HFETZMBQ"},"agent_actions":{"view_html":"https://pith.science/pith/HFETZMBQYJ4XW3ASQPFUAJE77A","download_json":"https://pith.science/pith/HFETZMBQYJ4XW3ASQPFUAJE77A.json","view_paper":"https://pith.science/paper/HFETZMBQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.05263&json=true","fetch_graph":"https://pith.science/api/pith-number/HFETZMBQYJ4XW3ASQPFUAJE77A/graph.json","fetch_events":"https://pith.science/api/pith-number/HFETZMBQYJ4XW3ASQPFUAJE77A/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HFETZMBQYJ4XW3ASQPFUAJE77A/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HFETZMBQYJ4XW3ASQPFUAJE77A/action/storage_attestation","attest_author":"https://pith.science/pith/HFETZMBQYJ4XW3ASQPFUAJE77A/action/author_attestation","sign_citation":"https://pith.science/pith/HFETZMBQYJ4XW3ASQPFUAJE77A/action/citation_signature","submit_replication":"https://pith.science/pith/HFETZMBQYJ4XW3ASQPFUAJE77A/action/replication_record"}},"created_at":"2026-08-07T00:46:38.171974+00:00","updated_at":"2026-08-07T00:46:38.171974+00:00"}