{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:4CWCNFARMZU4O6NJ3VRNHAZZZY","short_pith_number":"pith:4CWCNFAR","canonical_record":{"source":{"id":"2605.23701","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-22T14:52:32Z","cross_cats_sorted":[],"title_canon_sha256":"63f4f2325af8bbca71d9a5073552aca243ce107dde1930c3ffdce2e0a2dd46e2","abstract_canon_sha256":"cf36cfdc4dbdd265cdd153b1e34349d433487aeb8156d996a22782720865e549"},"schema_version":"1.0"},"canonical_sha256":"e0ac2694116669c779a9dd62d38339ce3f6ed6ad6aaa6f6766a9ee6469785a3c","source":{"kind":"arxiv","id":"2605.23701","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.23701","created_at":"2026-05-25T02:02:27Z"},{"alias_kind":"arxiv_version","alias_value":"2605.23701v1","created_at":"2026-05-25T02:02:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.23701","created_at":"2026-05-25T02:02:27Z"},{"alias_kind":"pith_short_12","alias_value":"4CWCNFARMZU4","created_at":"2026-05-25T02:02:27Z"},{"alias_kind":"pith_short_16","alias_value":"4CWCNFARMZU4O6NJ","created_at":"2026-05-25T02:02:27Z"},{"alias_kind":"pith_short_8","alias_value":"4CWCNFAR","created_at":"2026-05-25T02:02:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:4CWCNFARMZU4O6NJ3VRNHAZZZY","target":"record","payload":{"canonical_record":{"source":{"id":"2605.23701","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-22T14:52:32Z","cross_cats_sorted":[],"title_canon_sha256":"63f4f2325af8bbca71d9a5073552aca243ce107dde1930c3ffdce2e0a2dd46e2","abstract_canon_sha256":"cf36cfdc4dbdd265cdd153b1e34349d433487aeb8156d996a22782720865e549"},"schema_version":"1.0"},"canonical_sha256":"e0ac2694116669c779a9dd62d38339ce3f6ed6ad6aaa6f6766a9ee6469785a3c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-25T02:02:27.109263Z","signature_b64":"IAhtDj7raUV5Chco1SAsrth3Vsawrgv+nB8jcr96MtpHZsKedwYDZSMOcgQConLfvmYWH4imyFYUsAUU6k3vDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e0ac2694116669c779a9dd62d38339ce3f6ed6ad6aaa6f6766a9ee6469785a3c","last_reissued_at":"2026-05-25T02:02:27.108665Z","signature_status":"signed_v1","first_computed_at":"2026-05-25T02:02:27.108665Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.23701","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-25T02:02:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"o9EUvi2HAY2b0Ztc0hmn6nVFn+mEIDsVRiZ3yyj5pKt2DNJyj+cNLdhdviHePFZVKCXoUP+4rgj6yvvMcLixCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-10T08:21:25.150258Z"},"content_sha256":"e66e61406621fd97151669be8571919a66790ae6f0e9baa058e19fe645038598","schema_version":"1.0","event_id":"sha256:e66e61406621fd97151669be8571919a66790ae6f0e9baa058e19fe645038598"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:4CWCNFARMZU4O6NJ3VRNHAZZZY","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Metadata Predictability Is Not Evidence Dependence: An Intervention-Based Audit for Weak-Label Benchmarks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Kan Shao","submitted_at":"2026-05-22T14:52:32Z","abstract_excerpt":"We study a protocol-level test for weak-label benchmarks: whether benchmark outputs change when the provided evidence is intervened on. Metadata-only shortcut checks answer a different question, namely whether outputs are predictable from metadata priors. We therefore combine a metadata statistic, the Metadata Prior Dominance Score (MPDS), with an evidence-intervention statistic, {\\Delta}Evi, measuring sensitivity to evidence identity under cross-item shuffling. Synthetic HotpotQA gives a constructed counterexample to metadata-only screening: MPDS is only moderate (0.643), yet {\\Delta}Evi is z"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.23701","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/2605.23701/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":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-25T02:02:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZlF3euZd/jlHIjnZxg5Wg5m3cYL9jKBHHCpL8hKQ5IdrwyBq3E6dw4BnPoZxT8/tIvI79VmQHOHJSgYUAZvLCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-10T08:21:25.150948Z"},"content_sha256":"12a6ad9cbadc6e5f9eeb76cf3f3df71b80b140bd77ffeb61b0d1838196fcbaac","schema_version":"1.0","event_id":"sha256:12a6ad9cbadc6e5f9eeb76cf3f3df71b80b140bd77ffeb61b0d1838196fcbaac"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4CWCNFARMZU4O6NJ3VRNHAZZZY/bundle.json","state_url":"https://pith.science/pith/4CWCNFARMZU4O6NJ3VRNHAZZZY/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4CWCNFARMZU4O6NJ3VRNHAZZZY/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":"2026-06-10T08:21:25Z","links":{"resolver":"https://pith.science/pith/4CWCNFARMZU4O6NJ3VRNHAZZZY","bundle":"https://pith.science/pith/4CWCNFARMZU4O6NJ3VRNHAZZZY/bundle.json","state":"https://pith.science/pith/4CWCNFARMZU4O6NJ3VRNHAZZZY/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4CWCNFARMZU4O6NJ3VRNHAZZZY/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:4CWCNFARMZU4O6NJ3VRNHAZZZY","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":"cf36cfdc4dbdd265cdd153b1e34349d433487aeb8156d996a22782720865e549","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-22T14:52:32Z","title_canon_sha256":"63f4f2325af8bbca71d9a5073552aca243ce107dde1930c3ffdce2e0a2dd46e2"},"schema_version":"1.0","source":{"id":"2605.23701","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.23701","created_at":"2026-05-25T02:02:27Z"},{"alias_kind":"arxiv_version","alias_value":"2605.23701v1","created_at":"2026-05-25T02:02:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.23701","created_at":"2026-05-25T02:02:27Z"},{"alias_kind":"pith_short_12","alias_value":"4CWCNFARMZU4","created_at":"2026-05-25T02:02:27Z"},{"alias_kind":"pith_short_16","alias_value":"4CWCNFARMZU4O6NJ","created_at":"2026-05-25T02:02:27Z"},{"alias_kind":"pith_short_8","alias_value":"4CWCNFAR","created_at":"2026-05-25T02:02:27Z"}],"graph_snapshots":[{"event_id":"sha256:12a6ad9cbadc6e5f9eeb76cf3f3df71b80b140bd77ffeb61b0d1838196fcbaac","target":"graph","created_at":"2026-05-25T02:02:27Z","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":[],"endpoint":"/pith/2605.23701/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We study a protocol-level test for weak-label benchmarks: whether benchmark outputs change when the provided evidence is intervened on. Metadata-only shortcut checks answer a different question, namely whether outputs are predictable from metadata priors. We therefore combine a metadata statistic, the Metadata Prior Dominance Score (MPDS), with an evidence-intervention statistic, {\\Delta}Evi, measuring sensitivity to evidence identity under cross-item shuffling. Synthetic HotpotQA gives a constructed counterexample to metadata-only screening: MPDS is only moderate (0.643), yet {\\Delta}Evi is z","authors_text":"Kan Shao","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-22T14:52:32Z","title":"Metadata Predictability Is Not Evidence Dependence: An Intervention-Based Audit for Weak-Label Benchmarks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.23701","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:e66e61406621fd97151669be8571919a66790ae6f0e9baa058e19fe645038598","target":"record","created_at":"2026-05-25T02:02:27Z","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":"cf36cfdc4dbdd265cdd153b1e34349d433487aeb8156d996a22782720865e549","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-22T14:52:32Z","title_canon_sha256":"63f4f2325af8bbca71d9a5073552aca243ce107dde1930c3ffdce2e0a2dd46e2"},"schema_version":"1.0","source":{"id":"2605.23701","kind":"arxiv","version":1}},"canonical_sha256":"e0ac2694116669c779a9dd62d38339ce3f6ed6ad6aaa6f6766a9ee6469785a3c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e0ac2694116669c779a9dd62d38339ce3f6ed6ad6aaa6f6766a9ee6469785a3c","first_computed_at":"2026-05-25T02:02:27.108665Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-25T02:02:27.108665Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"IAhtDj7raUV5Chco1SAsrth3Vsawrgv+nB8jcr96MtpHZsKedwYDZSMOcgQConLfvmYWH4imyFYUsAUU6k3vDQ==","signature_status":"signed_v1","signed_at":"2026-05-25T02:02:27.109263Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.23701","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e66e61406621fd97151669be8571919a66790ae6f0e9baa058e19fe645038598","sha256:12a6ad9cbadc6e5f9eeb76cf3f3df71b80b140bd77ffeb61b0d1838196fcbaac"],"state_sha256":"6562d895b05fb84a40388a5707bbd1df403a704868e6036d6c632ae8ef91a16e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Y+0DXj8NBDwalD6UT6Ep1kPEFvqMHiaapdQDUNCwmiD7LkxM4P2fNhh6oVOginE7wmE2LLlOLwqrrYUuzUUxDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-10T08:21:25.155079Z","bundle_sha256":"46c7ea938ffa60c761cad93f084dc9e5b160e9883aeb2e9a64f078ed4bf680f0"}}