{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:WKYJLQMBAUVRUFRACMHMKGQWP4","short_pith_number":"pith:WKYJLQMB","canonical_record":{"source":{"id":"2604.14585","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-16T03:23:46Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"662f573db131e5cd7e7c7cd80bc165db3f7d6623bfbf3d3f4723bd87495016e4","abstract_canon_sha256":"ff912bb80d5464568b224fd1c42b15b927036bbb6e95114c81bb049248c97f08"},"schema_version":"1.0"},"canonical_sha256":"b2b095c181052b1a1620130ec51a167f28d185e94de65f8064e60be64198f54e","source":{"kind":"arxiv","id":"2604.14585","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.14585","created_at":"2026-05-28T01:04:40Z"},{"alias_kind":"arxiv_version","alias_value":"2604.14585v2","created_at":"2026-05-28T01:04:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.14585","created_at":"2026-05-28T01:04:40Z"},{"alias_kind":"pith_short_12","alias_value":"WKYJLQMBAUVR","created_at":"2026-05-28T01:04:40Z"},{"alias_kind":"pith_short_16","alias_value":"WKYJLQMBAUVRUFRA","created_at":"2026-05-28T01:04:40Z"},{"alias_kind":"pith_short_8","alias_value":"WKYJLQMB","created_at":"2026-05-28T01:04:40Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:WKYJLQMBAUVRUFRACMHMKGQWP4","target":"record","payload":{"canonical_record":{"source":{"id":"2604.14585","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-16T03:23:46Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"662f573db131e5cd7e7c7cd80bc165db3f7d6623bfbf3d3f4723bd87495016e4","abstract_canon_sha256":"ff912bb80d5464568b224fd1c42b15b927036bbb6e95114c81bb049248c97f08"},"schema_version":"1.0"},"canonical_sha256":"b2b095c181052b1a1620130ec51a167f28d185e94de65f8064e60be64198f54e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T01:04:40.145846Z","signature_b64":"/aPqtMu1Iv9BxTaKhBa7uO4LEWL3omUuDG0o/weGkJJ6OvuhS9ffs2FAjeJ2DVIwTS+17zRouN7mjo8PH5C2Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b2b095c181052b1a1620130ec51a167f28d185e94de65f8064e60be64198f54e","last_reissued_at":"2026-05-28T01:04:40.145303Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T01:04:40.145303Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2604.14585","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-28T01:04:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZFgNalkkEPQd1Jay3Lty1HYIHemWNuk44rsdQQ+/elQ+V0Dz5Pp0v7NpdHT9zuLOip1vx6nwM1kNfbA0gV5kAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-01T19:16:16.441925Z"},"content_sha256":"2e6a18d8f2d2e80c6fd424d7f046b193fd2104879f68cd1f149ed49a6229b909","schema_version":"1.0","event_id":"sha256:2e6a18d8f2d2e80c6fd424d7f046b193fd2104879f68cd1f149ed49a6229b909"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:WKYJLQMBAUVRUFRACMHMKGQWP4","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Prompt Optimization Is a Coin Flip: Diagnosing When It Helps in Compound AI Systems","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"Prompt optimization in compound AI systems performs no better than random chance on most tasks.","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Bing Zhu, Guanghui Wang, Peiyang He, Wei Qiu, Xing Zhang, Yanwei Cui, Ziyuan Li","submitted_at":"2026-04-16T03:23:46Z","abstract_excerpt":"Prompt optimization in compound AI systems is statistically indistinguishable from a coin flip: across 72 optimization runs on Claude Haiku 4.5 (6 methods $\\times$ 4 tasks $\\times$ 3 repeats), 49% score below zero-shot; on Amazon Nova Lite, the failure rate is even higher. Yet on one task, all six methods improve over zero-shot by up to $+6.8$ points. What distinguishes success from failure? We investigate with 18,000 grid evaluations and 144 optimization runs, testing two assumptions behind end-to-end optimization tools like TextGrad and DSPy, in the order they must be answered: (A) agent pro"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Prompt optimization in compound AI systems is statistically indistinguishable from a coin flip: across 72 optimization runs on Claude Haiku (6 methods × 4 tasks × 3 repeats), 49% score below zero-shot; ... optimization helps only when the task has exploitable output structure -- a format the model can produce but does not default to.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the tested tasks, models (Claude Haiku, Amazon Nova Lite), and optimization methods are representative enough for the conclusions about interaction effects (p > 0.52) and exploitable output structure to generalize to other compound AI systems.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Prompt optimization in compound AI systems is statistically indistinguishable from random chance except when tasks have exploitable output structure; a two-stage diagnostic predicts success.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Prompt optimization in compound AI systems performs no better than random chance on most tasks.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"e658e18bac694bde6c020690665d8a06dc1c5a191d1a25c8ad695aeeed33eb24"},"source":{"id":"2604.14585","kind":"arxiv","version":2},"verdict":{"id":"5b6170b0-f2ff-4fb6-9a54-d624623dd0fe","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T11:13:25.020340Z","strongest_claim":"Prompt optimization in compound AI systems is statistically indistinguishable from a coin flip: across 72 optimization runs on Claude Haiku (6 methods × 4 tasks × 3 repeats), 49% score below zero-shot; ... optimization helps only when the task has exploitable output structure -- a format the model can produce but does not default to.","one_line_summary":"Prompt optimization in compound AI systems is statistically indistinguishable from random chance except when tasks have exploitable output structure; a two-stage diagnostic predicts success.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the tested tasks, models (Claude Haiku, Amazon Nova Lite), and optimization methods are representative enough for the conclusions about interaction effects (p > 0.52) and exploitable output structure to generalize to other compound AI systems.","pith_extraction_headline":"Prompt optimization in compound AI systems performs no better than random chance on most tasks."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.14585/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"},"verdict_id":"5b6170b0-f2ff-4fb6-9a54-d624623dd0fe"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-28T01:04:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SCD7SbxaeBNey5WnZDcXADtF3MZYvbbaX2zE+UxrFpJveLt2THPoVO+GIOXLWeXSN02JACRVWN0LmFUi58OECg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-01T19:16:16.442413Z"},"content_sha256":"2158d40dca69f247f1ca85d58faeb6dcf6efee43bc1ad0b8a18eef510fa406de","schema_version":"1.0","event_id":"sha256:2158d40dca69f247f1ca85d58faeb6dcf6efee43bc1ad0b8a18eef510fa406de"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/WKYJLQMBAUVRUFRACMHMKGQWP4/bundle.json","state_url":"https://pith.science/pith/WKYJLQMBAUVRUFRACMHMKGQWP4/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/WKYJLQMBAUVRUFRACMHMKGQWP4/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-06-01T19:16:16Z","links":{"resolver":"https://pith.science/pith/WKYJLQMBAUVRUFRACMHMKGQWP4","bundle":"https://pith.science/pith/WKYJLQMBAUVRUFRACMHMKGQWP4/bundle.json","state":"https://pith.science/pith/WKYJLQMBAUVRUFRACMHMKGQWP4/state.json","well_known_bundle":"https://pith.science/.well-known/pith/WKYJLQMBAUVRUFRACMHMKGQWP4/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:WKYJLQMBAUVRUFRACMHMKGQWP4","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":"ff912bb80d5464568b224fd1c42b15b927036bbb6e95114c81bb049248c97f08","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-16T03:23:46Z","title_canon_sha256":"662f573db131e5cd7e7c7cd80bc165db3f7d6623bfbf3d3f4723bd87495016e4"},"schema_version":"1.0","source":{"id":"2604.14585","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.14585","created_at":"2026-05-28T01:04:40Z"},{"alias_kind":"arxiv_version","alias_value":"2604.14585v2","created_at":"2026-05-28T01:04:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.14585","created_at":"2026-05-28T01:04:40Z"},{"alias_kind":"pith_short_12","alias_value":"WKYJLQMBAUVR","created_at":"2026-05-28T01:04:40Z"},{"alias_kind":"pith_short_16","alias_value":"WKYJLQMBAUVRUFRA","created_at":"2026-05-28T01:04:40Z"},{"alias_kind":"pith_short_8","alias_value":"WKYJLQMB","created_at":"2026-05-28T01:04:40Z"}],"graph_snapshots":[{"event_id":"sha256:2158d40dca69f247f1ca85d58faeb6dcf6efee43bc1ad0b8a18eef510fa406de","target":"graph","created_at":"2026-05-28T01:04:40Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Prompt optimization in compound AI systems is statistically indistinguishable from a coin flip: across 72 optimization runs on Claude Haiku (6 methods × 4 tasks × 3 repeats), 49% score below zero-shot; ... optimization helps only when the task has exploitable output structure -- a format the model can produce but does not default to."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the tested tasks, models (Claude Haiku, Amazon Nova Lite), and optimization methods are representative enough for the conclusions about interaction effects (p > 0.52) and exploitable output structure to generalize to other compound AI systems."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Prompt optimization in compound AI systems is statistically indistinguishable from random chance except when tasks have exploitable output structure; a two-stage diagnostic predicts success."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Prompt optimization in compound AI systems performs no better than random chance on most tasks."}],"snapshot_sha256":"e658e18bac694bde6c020690665d8a06dc1c5a191d1a25c8ad695aeeed33eb24"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2604.14585/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Prompt optimization in compound AI systems is statistically indistinguishable from a coin flip: across 72 optimization runs on Claude Haiku 4.5 (6 methods $\\times$ 4 tasks $\\times$ 3 repeats), 49% score below zero-shot; on Amazon Nova Lite, the failure rate is even higher. Yet on one task, all six methods improve over zero-shot by up to $+6.8$ points. What distinguishes success from failure? We investigate with 18,000 grid evaluations and 144 optimization runs, testing two assumptions behind end-to-end optimization tools like TextGrad and DSPy, in the order they must be answered: (A) agent pro","authors_text":"Bing Zhu, Guanghui Wang, Peiyang He, Wei Qiu, Xing Zhang, Yanwei Cui, Ziyuan Li","cross_cats":["cs.CL"],"headline":"Prompt optimization in compound AI systems performs no better than random chance on most tasks.","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-16T03:23:46Z","title":"Prompt Optimization Is a Coin Flip: Diagnosing When It Helps in Compound AI Systems"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.14585","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-10T11:13:25.020340Z","id":"5b6170b0-f2ff-4fb6-9a54-d624623dd0fe","model_set":{"reader":"grok-4.3"},"one_line_summary":"Prompt optimization in compound AI systems is statistically indistinguishable from random chance except when tasks have exploitable output structure; a two-stage diagnostic predicts success.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Prompt optimization in compound AI systems performs no better than random chance on most tasks.","strongest_claim":"Prompt optimization in compound AI systems is statistically indistinguishable from a coin flip: across 72 optimization runs on Claude Haiku (6 methods × 4 tasks × 3 repeats), 49% score below zero-shot; ... optimization helps only when the task has exploitable output structure -- a format the model can produce but does not default to.","weakest_assumption":"That the tested tasks, models (Claude Haiku, Amazon Nova Lite), and optimization methods are representative enough for the conclusions about interaction effects (p > 0.52) and exploitable output structure to generalize to other compound AI systems."}},"verdict_id":"5b6170b0-f2ff-4fb6-9a54-d624623dd0fe"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:2e6a18d8f2d2e80c6fd424d7f046b193fd2104879f68cd1f149ed49a6229b909","target":"record","created_at":"2026-05-28T01:04:40Z","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":"ff912bb80d5464568b224fd1c42b15b927036bbb6e95114c81bb049248c97f08","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-16T03:23:46Z","title_canon_sha256":"662f573db131e5cd7e7c7cd80bc165db3f7d6623bfbf3d3f4723bd87495016e4"},"schema_version":"1.0","source":{"id":"2604.14585","kind":"arxiv","version":2}},"canonical_sha256":"b2b095c181052b1a1620130ec51a167f28d185e94de65f8064e60be64198f54e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b2b095c181052b1a1620130ec51a167f28d185e94de65f8064e60be64198f54e","first_computed_at":"2026-05-28T01:04:40.145303Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-28T01:04:40.145303Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"/aPqtMu1Iv9BxTaKhBa7uO4LEWL3omUuDG0o/weGkJJ6OvuhS9ffs2FAjeJ2DVIwTS+17zRouN7mjo8PH5C2Dg==","signature_status":"signed_v1","signed_at":"2026-05-28T01:04:40.145846Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.14585","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2e6a18d8f2d2e80c6fd424d7f046b193fd2104879f68cd1f149ed49a6229b909","sha256:2158d40dca69f247f1ca85d58faeb6dcf6efee43bc1ad0b8a18eef510fa406de"],"state_sha256":"081580115620193e263402c58a2f5ae944351a9833558c07a7389d82ec810698"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"L4KkaVp+rOHYXM7csZB/fgsuDjHCuCBj/GCgZLVf6azdKCZTIAJ/ShP1iW4aL0uNcQ+DsIVykr8KnnmGaRQ2BA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-01T19:16:16.444793Z","bundle_sha256":"e5d80597549ffe377e2360aac975d7a3455e6e1534dbd5804df77626eed0839b"}}