{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:QNGUHUXXFFVZRZO7AQJCB6KFBH","short_pith_number":"pith:QNGUHUXX","canonical_record":{"source":{"id":"2607.24268","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-27T11:04:17Z","cross_cats_sorted":[],"title_canon_sha256":"a2a1d02c09676596a1d3a9c2b015e15a9f7b19c3aa8bf06d58b643d6112e939e","abstract_canon_sha256":"11298b943a7ed982a4cf01d6226055ffc3b7a4ce7ad95f647f11f19b9d75fa00"},"schema_version":"1.0"},"canonical_sha256":"834d43d2f7296b98e5df041220f94509cbf77538d9a0efa504dde444e42a9762","source":{"kind":"arxiv","id":"2607.24268","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.24268","created_at":"2026-07-28T02:23:54Z"},{"alias_kind":"arxiv_version","alias_value":"2607.24268v1","created_at":"2026-07-28T02:23:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.24268","created_at":"2026-07-28T02:23:54Z"},{"alias_kind":"pith_short_12","alias_value":"QNGUHUXXFFVZ","created_at":"2026-07-28T02:23:54Z"},{"alias_kind":"pith_short_16","alias_value":"QNGUHUXXFFVZRZO7","created_at":"2026-07-28T02:23:54Z"},{"alias_kind":"pith_short_8","alias_value":"QNGUHUXX","created_at":"2026-07-28T02:23:54Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:QNGUHUXXFFVZRZO7AQJCB6KFBH","target":"record","payload":{"canonical_record":{"source":{"id":"2607.24268","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-27T11:04:17Z","cross_cats_sorted":[],"title_canon_sha256":"a2a1d02c09676596a1d3a9c2b015e15a9f7b19c3aa8bf06d58b643d6112e939e","abstract_canon_sha256":"11298b943a7ed982a4cf01d6226055ffc3b7a4ce7ad95f647f11f19b9d75fa00"},"schema_version":"1.0"},"canonical_sha256":"834d43d2f7296b98e5df041220f94509cbf77538d9a0efa504dde444e42a9762","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-28T02:23:54.475923Z","signature_b64":"cz0vmkMJTaNMXdG+kTEg5g/HjLNlcmKY7V4/ZQipnKffg+IRTI6bEcKQWMZ9l3AHrkHuLwHid2vF6qee6CjJCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"834d43d2f7296b98e5df041220f94509cbf77538d9a0efa504dde444e42a9762","last_reissued_at":"2026-07-28T02:23:54.475115Z","signature_status":"signed_v1","first_computed_at":"2026-07-28T02:23:54.475115Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.24268","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":"2026-07-28T02:23:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SR30N14Us5NY+t+4vjk/hFpHSeF9+sLl0antkmQtJnQgyTmmPiqq/yTHAevIIV1lhJe1gWEEXAVLcYp1BkDCCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T15:47:41.123334Z"},"content_sha256":"db35c529f67431486f13852842cd8274f07a6e68fb6411c29b773d2c59505f99","schema_version":"1.0","event_id":"sha256:db35c529f67431486f13852842cd8274f07a6e68fb6411c29b773d2c59505f99"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:QNGUHUXXFFVZRZO7AQJCB6KFBH","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Accuracy Hides How Language Models Fail: Measuring Failure States Under Matched Output Budgets","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Yinghan Hou, Zongyou Yang","submitted_at":"2026-07-27T11:04:17Z","abstract_excerpt":"Language-model benchmarks collapse two distinct measurement questions into a single accuracy score: whether a response reached an evaluable state, and whether its answer was judged correct. We introduce a two-layer evaluation framework that separates scorer-independent execution evidence, including termination, answer exposure, parseability, and completion length, from scorer-dependent correctness. Across 2,550 outputs from five fixed Qwen and DeepSeek configurations on MATH and ARC-Challenge, matched 2,048-token limits produce sharply different execution mixtures: 49 of 450 Qwen MATH outputs "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.24268","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/2607.24268/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":"2026-07-28T02:23:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"EePHOH3lRbByUxycxCfGSoNblL6U716BPvi/EUtND+INrgcKUmjGL4XDYLvJql0Sb0JY12dacJ7OS51cKcMsCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T15:47:41.124321Z"},"content_sha256":"78b68f68ad387802f0f80705ddf894a8f494f6b6d54e814425cda16a69f14a49","schema_version":"1.0","event_id":"sha256:78b68f68ad387802f0f80705ddf894a8f494f6b6d54e814425cda16a69f14a49"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QNGUHUXXFFVZRZO7AQJCB6KFBH/bundle.json","state_url":"https://pith.science/pith/QNGUHUXXFFVZRZO7AQJCB6KFBH/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QNGUHUXXFFVZRZO7AQJCB6KFBH/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":"2026-08-18T15:47:41Z","links":{"resolver":"https://pith.science/pith/QNGUHUXXFFVZRZO7AQJCB6KFBH","bundle":"https://pith.science/pith/QNGUHUXXFFVZRZO7AQJCB6KFBH/bundle.json","state":"https://pith.science/pith/QNGUHUXXFFVZRZO7AQJCB6KFBH/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QNGUHUXXFFVZRZO7AQJCB6KFBH/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:QNGUHUXXFFVZRZO7AQJCB6KFBH","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":"11298b943a7ed982a4cf01d6226055ffc3b7a4ce7ad95f647f11f19b9d75fa00","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-27T11:04:17Z","title_canon_sha256":"a2a1d02c09676596a1d3a9c2b015e15a9f7b19c3aa8bf06d58b643d6112e939e"},"schema_version":"1.0","source":{"id":"2607.24268","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.24268","created_at":"2026-07-28T02:23:54Z"},{"alias_kind":"arxiv_version","alias_value":"2607.24268v1","created_at":"2026-07-28T02:23:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.24268","created_at":"2026-07-28T02:23:54Z"},{"alias_kind":"pith_short_12","alias_value":"QNGUHUXXFFVZ","created_at":"2026-07-28T02:23:54Z"},{"alias_kind":"pith_short_16","alias_value":"QNGUHUXXFFVZRZO7","created_at":"2026-07-28T02:23:54Z"},{"alias_kind":"pith_short_8","alias_value":"QNGUHUXX","created_at":"2026-07-28T02:23:54Z"}],"graph_snapshots":[{"event_id":"sha256:78b68f68ad387802f0f80705ddf894a8f494f6b6d54e814425cda16a69f14a49","target":"graph","created_at":"2026-07-28T02:23:54Z","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/2607.24268/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Language-model benchmarks collapse two distinct measurement questions into a single accuracy score: whether a response reached an evaluable state, and whether its answer was judged correct. We introduce a two-layer evaluation framework that separates scorer-independent execution evidence, including termination, answer exposure, parseability, and completion length, from scorer-dependent correctness. Across 2,550 outputs from five fixed Qwen and DeepSeek configurations on MATH and ARC-Challenge, matched 2,048-token limits produce sharply different execution mixtures: 49 of 450 Qwen MATH outputs ","authors_text":"Yinghan Hou, Zongyou Yang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-27T11:04:17Z","title":"Accuracy Hides How Language Models Fail: Measuring Failure States Under Matched Output Budgets"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.24268","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:db35c529f67431486f13852842cd8274f07a6e68fb6411c29b773d2c59505f99","target":"record","created_at":"2026-07-28T02:23:54Z","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":"11298b943a7ed982a4cf01d6226055ffc3b7a4ce7ad95f647f11f19b9d75fa00","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-27T11:04:17Z","title_canon_sha256":"a2a1d02c09676596a1d3a9c2b015e15a9f7b19c3aa8bf06d58b643d6112e939e"},"schema_version":"1.0","source":{"id":"2607.24268","kind":"arxiv","version":1}},"canonical_sha256":"834d43d2f7296b98e5df041220f94509cbf77538d9a0efa504dde444e42a9762","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"834d43d2f7296b98e5df041220f94509cbf77538d9a0efa504dde444e42a9762","first_computed_at":"2026-07-28T02:23:54.475115Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-28T02:23:54.475115Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"cz0vmkMJTaNMXdG+kTEg5g/HjLNlcmKY7V4/ZQipnKffg+IRTI6bEcKQWMZ9l3AHrkHuLwHid2vF6qee6CjJCA==","signature_status":"signed_v1","signed_at":"2026-07-28T02:23:54.475923Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.24268","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:db35c529f67431486f13852842cd8274f07a6e68fb6411c29b773d2c59505f99","sha256:78b68f68ad387802f0f80705ddf894a8f494f6b6d54e814425cda16a69f14a49"],"state_sha256":"3b247ebc052dc0bef61a60cc9b6688ccbf1a7725a7d1d415a31b0620edbc6e5f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TsqOROUTTm63i4fo94Tz9Defiwet8fzG00leO2Xo7bV82GbMMSxFDyolNiUdh0FyRhJs1TO9QHdb5562UYI/DQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-18T15:47:41.130584Z","bundle_sha256":"1348b5da07ccb0ac9400296086d38691a79f16088ad3a234ae3d79ea8753da16"}}