{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:EDUYBA7ID62HMQU3EGSN2M7OZI","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"0b2815276788cadb5698bfae15192d84be7208f68469ea08a3c847fceb0ff1ed","cross_cats_sorted":["cs.AI","cs.HC","cs.SE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-18T08:22:14Z","title_canon_sha256":"446bc734039ce1aa652cea99e39450d5a25ca3e7a99f6deb32df01ed51ee63a8"},"schema_version":"1.0","source":{"id":"2605.18032","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.18032","created_at":"2026-05-20T00:05:12Z"},{"alias_kind":"arxiv_version","alias_value":"2605.18032v1","created_at":"2026-05-20T00:05:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.18032","created_at":"2026-05-20T00:05:12Z"},{"alias_kind":"pith_short_12","alias_value":"EDUYBA7ID62H","created_at":"2026-05-20T00:05:12Z"},{"alias_kind":"pith_short_16","alias_value":"EDUYBA7ID62HMQU3","created_at":"2026-05-20T00:05:12Z"},{"alias_kind":"pith_short_8","alias_value":"EDUYBA7I","created_at":"2026-05-20T00:05:12Z"}],"graph_snapshots":[{"event_id":"sha256:4adca0ee9f68ac23b64221b75a90ff0a9af69f18c062a06b2058fcfd19c46169","target":"graph","created_at":"2026-05-20T00:05:12Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T23:33:35.506936Z","status":"skipped","version":"1.0.0"}],"endpoint":"/pith/2605.18032/integrity.json","findings":[],"snapshot_sha256":"4c1e03acfcacbe9a2b6d0f1a8489d5714b8029b2037872d34cdb255c3cbf6bf3","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multi-agent LLM workflows -- systems composed of multiple role-specific LLM calls -- often outperform single-prompt baselines, but they remain difficult to debug and refine. Failures can originate from subtle errors in intermediate outputs that propagate to downstream nodes, requiring developers to inspect long traces and infer which agent to modify. We present PROTEA, a unified interface for offline, test-driven improvement of multi-agent workflows. PROTEA executes a workflow, scores intermediate node outputs with configurable rubrics, and overlays per-node states and rationales on the workfl","authors_text":"Kazuki Kawamura, Kei Tateno, Satoshi Waki","cross_cats":["cs.AI","cs.HC","cs.SE"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-18T08:22:14Z","title":"PROTEA: Offline Evaluation and Iterative Refinement for Multi-Agent LLM Workflows"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.18032","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:b1be50ae17552ffe8ad43ccfb81128e0be6346605788ef908fb7a22a87e45b41","target":"record","created_at":"2026-05-20T00:05:12Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"0b2815276788cadb5698bfae15192d84be7208f68469ea08a3c847fceb0ff1ed","cross_cats_sorted":["cs.AI","cs.HC","cs.SE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-18T08:22:14Z","title_canon_sha256":"446bc734039ce1aa652cea99e39450d5a25ca3e7a99f6deb32df01ed51ee63a8"},"schema_version":"1.0","source":{"id":"2605.18032","kind":"arxiv","version":1}},"canonical_sha256":"20e98083e81fb476429b21a4dd33eeca2d9aba9acac45fd90d893816edae35ec","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"20e98083e81fb476429b21a4dd33eeca2d9aba9acac45fd90d893816edae35ec","first_computed_at":"2026-05-20T00:05:12.266747Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:05:12.266747Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"QVt2tw9dnXO+fX8P/DCI3vKRfuaLgJjU5/mAv+VumKWiZZ0Jy8yIsY0Z0CNUL6XuL4oO16Gz8D8LrzkwcpALDw==","signature_status":"signed_v1","signed_at":"2026-05-20T00:05:12.267507Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.18032","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b1be50ae17552ffe8ad43ccfb81128e0be6346605788ef908fb7a22a87e45b41","sha256:4adca0ee9f68ac23b64221b75a90ff0a9af69f18c062a06b2058fcfd19c46169"],"state_sha256":"6a01e0da91762d8dfc305768e9a6a4ff2c5ef6e16141e2374f84912b9dc47b3f"}