{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:XLVN67C3QJT2DGDPZGJZB42XD6","short_pith_number":"pith:XLVN67C3","canonical_record":{"source":{"id":"2502.20747","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.SE","submitted_at":"2025-02-28T05:53:16Z","cross_cats_sorted":[],"title_canon_sha256":"0e3d0ef6863164743d7e77d72aab374333ff9dd2de412d4b0faf7ef988cbb637","abstract_canon_sha256":"be022d6a23d215b706de76fad3bd942773b411a6aaae4cfcf169fd686837ea1f"},"schema_version":"1.0"},"canonical_sha256":"baeadf7c5b8267a1986fc99390f3571fa4f8c7e109834f0d268e4b3ad8bf29d3","source":{"kind":"arxiv","id":"2502.20747","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.20747","created_at":"2026-07-05T10:21:38Z"},{"alias_kind":"arxiv_version","alias_value":"2502.20747v1","created_at":"2026-07-05T10:21:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.20747","created_at":"2026-07-05T10:21:38Z"},{"alias_kind":"pith_short_12","alias_value":"XLVN67C3QJT2","created_at":"2026-07-05T10:21:38Z"},{"alias_kind":"pith_short_16","alias_value":"XLVN67C3QJT2DGDP","created_at":"2026-07-05T10:21:38Z"},{"alias_kind":"pith_short_8","alias_value":"XLVN67C3","created_at":"2026-07-05T10:21:38Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:XLVN67C3QJT2DGDPZGJZB42XD6","target":"record","payload":{"canonical_record":{"source":{"id":"2502.20747","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.SE","submitted_at":"2025-02-28T05:53:16Z","cross_cats_sorted":[],"title_canon_sha256":"0e3d0ef6863164743d7e77d72aab374333ff9dd2de412d4b0faf7ef988cbb637","abstract_canon_sha256":"be022d6a23d215b706de76fad3bd942773b411a6aaae4cfcf169fd686837ea1f"},"schema_version":"1.0"},"canonical_sha256":"baeadf7c5b8267a1986fc99390f3571fa4f8c7e109834f0d268e4b3ad8bf29d3","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:21:38.974219Z","signature_b64":"NxDERLDyCkmfXI66KcknMPlhMsBRMFt9vWHdSzlX7bD7isyx2QIyr2mGqIYdF60L+oLQ4rmb3Smy1dvHO4ueAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"baeadf7c5b8267a1986fc99390f3571fa4f8c7e109834f0d268e4b3ad8bf29d3","last_reissued_at":"2026-07-05T10:21:38.973716Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:21:38.973716Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.20747","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":"2026-07-05T10:21:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"l9vHKxwEc+wdyyYiqmJzBM+5nFhS/lK62Xp+S1aITGx8SfETMwFwzGDvWHr+V4TxExDq8UFm9jZ6jY0dxXKeAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T00:20:07.632015Z"},"content_sha256":"086209334ea876d9a6c8931415072784aef4815a760f101a73f0726c7479ab07","schema_version":"1.0","event_id":"sha256:086209334ea876d9a6c8931415072784aef4815a760f101a73f0726c7479ab07"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:XLVN67C3QJT2DGDPZGJZB42XD6","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Measuring Determinism in Large Language Models for Software Code Review","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.SE","authors_text":"Eugene Klishevich, Igor Ciobanu, Michal Kosinski, Simon Obstbaum, Yegor Denisov-Blanch","submitted_at":"2025-02-28T05:53:16Z","abstract_excerpt":"Large Language Models (LLMs) promise to streamline software code reviews, but their ability to produce consistent assessments remains an open question. In this study, we tested four leading LLMs -- GPT-4o mini, GPT-4o, Claude 3.5 Sonnet, and LLaMA 3.2 90B Vision -- on 70 Java commits from both private and public repositories. By setting each model's temperature to zero, clearing context, and repeating the exact same prompts five times, we measured how consistently each model generated code-review assessments. Our results reveal that even with temperature minimized, LLM responses varied to diff"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.20747","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/2502.20747/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":"2026-07-05T10:21:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Rs/lpS5DNoMS/N4rbwOHQgyuBPuXaVdimEgkqRGYL+kokO0faQdeDtHMif1gMH7KCak4BZUCOFcWIpZ6gcSeAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T00:20:07.633027Z"},"content_sha256":"cc0808015c3a5a86e45847bd9b6af0c7bd6800ca64b8741665b72a3d27540696","schema_version":"1.0","event_id":"sha256:cc0808015c3a5a86e45847bd9b6af0c7bd6800ca64b8741665b72a3d27540696"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XLVN67C3QJT2DGDPZGJZB42XD6/bundle.json","state_url":"https://pith.science/pith/XLVN67C3QJT2DGDPZGJZB42XD6/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XLVN67C3QJT2DGDPZGJZB42XD6/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":"2026-08-09T00:20:07Z","links":{"resolver":"https://pith.science/pith/XLVN67C3QJT2DGDPZGJZB42XD6","bundle":"https://pith.science/pith/XLVN67C3QJT2DGDPZGJZB42XD6/bundle.json","state":"https://pith.science/pith/XLVN67C3QJT2DGDPZGJZB42XD6/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XLVN67C3QJT2DGDPZGJZB42XD6/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:XLVN67C3QJT2DGDPZGJZB42XD6","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":"be022d6a23d215b706de76fad3bd942773b411a6aaae4cfcf169fd686837ea1f","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.SE","submitted_at":"2025-02-28T05:53:16Z","title_canon_sha256":"0e3d0ef6863164743d7e77d72aab374333ff9dd2de412d4b0faf7ef988cbb637"},"schema_version":"1.0","source":{"id":"2502.20747","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.20747","created_at":"2026-07-05T10:21:38Z"},{"alias_kind":"arxiv_version","alias_value":"2502.20747v1","created_at":"2026-07-05T10:21:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.20747","created_at":"2026-07-05T10:21:38Z"},{"alias_kind":"pith_short_12","alias_value":"XLVN67C3QJT2","created_at":"2026-07-05T10:21:38Z"},{"alias_kind":"pith_short_16","alias_value":"XLVN67C3QJT2DGDP","created_at":"2026-07-05T10:21:38Z"},{"alias_kind":"pith_short_8","alias_value":"XLVN67C3","created_at":"2026-07-05T10:21:38Z"}],"graph_snapshots":[{"event_id":"sha256:cc0808015c3a5a86e45847bd9b6af0c7bd6800ca64b8741665b72a3d27540696","target":"graph","created_at":"2026-07-05T10:21:38Z","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/2502.20747/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) promise to streamline software code reviews, but their ability to produce consistent assessments remains an open question. In this study, we tested four leading LLMs -- GPT-4o mini, GPT-4o, Claude 3.5 Sonnet, and LLaMA 3.2 90B Vision -- on 70 Java commits from both private and public repositories. By setting each model's temperature to zero, clearing context, and repeating the exact same prompts five times, we measured how consistently each model generated code-review assessments. Our results reveal that even with temperature minimized, LLM responses varied to diff","authors_text":"Eugene Klishevich, Igor Ciobanu, Michal Kosinski, Simon Obstbaum, Yegor Denisov-Blanch","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.SE","submitted_at":"2025-02-28T05:53:16Z","title":"Measuring Determinism in Large Language Models for Software Code Review"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.20747","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":"sha256:086209334ea876d9a6c8931415072784aef4815a760f101a73f0726c7479ab07","target":"record","created_at":"2026-07-05T10:21:38Z","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":"be022d6a23d215b706de76fad3bd942773b411a6aaae4cfcf169fd686837ea1f","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.SE","submitted_at":"2025-02-28T05:53:16Z","title_canon_sha256":"0e3d0ef6863164743d7e77d72aab374333ff9dd2de412d4b0faf7ef988cbb637"},"schema_version":"1.0","source":{"id":"2502.20747","kind":"arxiv","version":1}},"canonical_sha256":"baeadf7c5b8267a1986fc99390f3571fa4f8c7e109834f0d268e4b3ad8bf29d3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"baeadf7c5b8267a1986fc99390f3571fa4f8c7e109834f0d268e4b3ad8bf29d3","first_computed_at":"2026-07-05T10:21:38.973716Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:21:38.973716Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"NxDERLDyCkmfXI66KcknMPlhMsBRMFt9vWHdSzlX7bD7isyx2QIyr2mGqIYdF60L+oLQ4rmb3Smy1dvHO4ueAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:21:38.974219Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.20747","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:086209334ea876d9a6c8931415072784aef4815a760f101a73f0726c7479ab07","sha256:cc0808015c3a5a86e45847bd9b6af0c7bd6800ca64b8741665b72a3d27540696"],"state_sha256":"b89e5289381b1e108a193c62d99cef1b7c909baffb9543902073367498e951c4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TkfFqG7sWnuy4vIIqkugqvbWeNkLTSTj+gEfAPWj++ZeCiuFMukcVxBkDVbiHcWLQYWb4yiBdjvy1dlbCTEFCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T00:20:07.637753Z","bundle_sha256":"52ff19a3fb792da3a5504cb247a12b6af9e070b7c93738568bcee1f1a4133bdc"}}