{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:U6W6SM2EGP7GV4PFWLJ4BMXYD5","short_pith_number":"pith:U6W6SM2E","canonical_record":{"source":{"id":"2509.10963","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.ST","submitted_at":"2025-09-13T19:44:42Z","cross_cats_sorted":["cs.AI","stat.ME","stat.TH"],"title_canon_sha256":"2e4dd5f6c514e3b1b3111cb79f4760dc73e501b14a0defa6db92744b18afc3eb","abstract_canon_sha256":"cb150c805a02c51904cbd5ddf1e8cfbbca80a0136b178ec643684a41ee69f023"},"schema_version":"1.0"},"canonical_sha256":"a7ade9334433fe6af1e5b2d3c0b2f81f5788681528dee8c532ad736535870d03","source":{"kind":"arxiv","id":"2509.10963","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.10963","created_at":"2026-07-05T12:11:31Z"},{"alias_kind":"arxiv_version","alias_value":"2509.10963v1","created_at":"2026-07-05T12:11:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.10963","created_at":"2026-07-05T12:11:31Z"},{"alias_kind":"pith_short_12","alias_value":"U6W6SM2EGP7G","created_at":"2026-07-05T12:11:31Z"},{"alias_kind":"pith_short_16","alias_value":"U6W6SM2EGP7GV4PF","created_at":"2026-07-05T12:11:31Z"},{"alias_kind":"pith_short_8","alias_value":"U6W6SM2E","created_at":"2026-07-05T12:11:31Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:U6W6SM2EGP7GV4PFWLJ4BMXYD5","target":"record","payload":{"canonical_record":{"source":{"id":"2509.10963","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.ST","submitted_at":"2025-09-13T19:44:42Z","cross_cats_sorted":["cs.AI","stat.ME","stat.TH"],"title_canon_sha256":"2e4dd5f6c514e3b1b3111cb79f4760dc73e501b14a0defa6db92744b18afc3eb","abstract_canon_sha256":"cb150c805a02c51904cbd5ddf1e8cfbbca80a0136b178ec643684a41ee69f023"},"schema_version":"1.0"},"canonical_sha256":"a7ade9334433fe6af1e5b2d3c0b2f81f5788681528dee8c532ad736535870d03","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:11:31.204519Z","signature_b64":"iNTzht8h0XlpF03fHA2k6cgkcazMxh+hZLyahDVkpT2C5TvLD/jONyfHi7Ve3BfltWS/PebeZVj++guYtleLBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a7ade9334433fe6af1e5b2d3c0b2f81f5788681528dee8c532ad736535870d03","last_reissued_at":"2026-07-05T12:11:31.203985Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:11:31.203985Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2509.10963","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":"2026-07-05T12:11:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tG/ewi8kbORB1xiYK4G7L1iZbbvfjZTCAorilLxGd/TLofDvWUbc8w+fP9xqaUKdcEWAn3060hbwS8GFucQhCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-27T02:26:56.996107Z"},"content_sha256":"a27f99814a1ff7d20e6b4f82829e137ae81509934137f63113f5a16e50448a5d","schema_version":"1.0","event_id":"sha256:a27f99814a1ff7d20e6b4f82829e137ae81509934137f63113f5a16e50448a5d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:U6W6SM2EGP7GV4PFWLJ4BMXYD5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Testing for LLM response differences: the case of a composite null consisting of semantically irrelevant query perturbations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ME","stat.TH"],"primary_cat":"math.ST","authors_text":"Aranyak Acharyya, Carey E. Priebe, Hayden S. Helm","submitted_at":"2025-09-13T19:44:42Z","abstract_excerpt":"Given an input query, generative models such as large language models produce a random response drawn from a response distribution. Given two input queries, it is natural to ask if their response distributions are the same. While traditional statistical hypothesis testing is designed to address this question, the response distribution induced by an input query is often sensitive to semantically irrelevant perturbations to the query, so much so that a traditional test of equality might indicate that two semantically equivalent queries induce statistically different response distributions. As a "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.10963","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/2509.10963/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":"2026-07-05T12:11:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uUwl1lpftMuRf0C5PyHAt2NZGNyDVRUGyi58DAPT93CScizisIrDIwUAjEQNfSEbRBeXa2n9USGI8TOU40ofAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-27T02:26:56.996500Z"},"content_sha256":"83aa9f726abbc5cb13c3a25c76e418b602a9e0da1ab21c23c70e8d4ab497644e","schema_version":"1.0","event_id":"sha256:83aa9f726abbc5cb13c3a25c76e418b602a9e0da1ab21c23c70e8d4ab497644e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/U6W6SM2EGP7GV4PFWLJ4BMXYD5/bundle.json","state_url":"https://pith.science/pith/U6W6SM2EGP7GV4PFWLJ4BMXYD5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/U6W6SM2EGP7GV4PFWLJ4BMXYD5/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":"2026-07-27T02:26:56Z","links":{"resolver":"https://pith.science/pith/U6W6SM2EGP7GV4PFWLJ4BMXYD5","bundle":"https://pith.science/pith/U6W6SM2EGP7GV4PFWLJ4BMXYD5/bundle.json","state":"https://pith.science/pith/U6W6SM2EGP7GV4PFWLJ4BMXYD5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/U6W6SM2EGP7GV4PFWLJ4BMXYD5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:U6W6SM2EGP7GV4PFWLJ4BMXYD5","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":"cb150c805a02c51904cbd5ddf1e8cfbbca80a0136b178ec643684a41ee69f023","cross_cats_sorted":["cs.AI","stat.ME","stat.TH"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.ST","submitted_at":"2025-09-13T19:44:42Z","title_canon_sha256":"2e4dd5f6c514e3b1b3111cb79f4760dc73e501b14a0defa6db92744b18afc3eb"},"schema_version":"1.0","source":{"id":"2509.10963","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.10963","created_at":"2026-07-05T12:11:31Z"},{"alias_kind":"arxiv_version","alias_value":"2509.10963v1","created_at":"2026-07-05T12:11:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.10963","created_at":"2026-07-05T12:11:31Z"},{"alias_kind":"pith_short_12","alias_value":"U6W6SM2EGP7G","created_at":"2026-07-05T12:11:31Z"},{"alias_kind":"pith_short_16","alias_value":"U6W6SM2EGP7GV4PF","created_at":"2026-07-05T12:11:31Z"},{"alias_kind":"pith_short_8","alias_value":"U6W6SM2E","created_at":"2026-07-05T12:11:31Z"}],"graph_snapshots":[{"event_id":"sha256:83aa9f726abbc5cb13c3a25c76e418b602a9e0da1ab21c23c70e8d4ab497644e","target":"graph","created_at":"2026-07-05T12:11:31Z","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/2509.10963/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Given an input query, generative models such as large language models produce a random response drawn from a response distribution. Given two input queries, it is natural to ask if their response distributions are the same. While traditional statistical hypothesis testing is designed to address this question, the response distribution induced by an input query is often sensitive to semantically irrelevant perturbations to the query, so much so that a traditional test of equality might indicate that two semantically equivalent queries induce statistically different response distributions. As a ","authors_text":"Aranyak Acharyya, Carey E. Priebe, Hayden S. Helm","cross_cats":["cs.AI","stat.ME","stat.TH"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.ST","submitted_at":"2025-09-13T19:44:42Z","title":"Testing for LLM response differences: the case of a composite null consisting of semantically irrelevant query perturbations"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.10963","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:a27f99814a1ff7d20e6b4f82829e137ae81509934137f63113f5a16e50448a5d","target":"record","created_at":"2026-07-05T12:11:31Z","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":"cb150c805a02c51904cbd5ddf1e8cfbbca80a0136b178ec643684a41ee69f023","cross_cats_sorted":["cs.AI","stat.ME","stat.TH"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.ST","submitted_at":"2025-09-13T19:44:42Z","title_canon_sha256":"2e4dd5f6c514e3b1b3111cb79f4760dc73e501b14a0defa6db92744b18afc3eb"},"schema_version":"1.0","source":{"id":"2509.10963","kind":"arxiv","version":1}},"canonical_sha256":"a7ade9334433fe6af1e5b2d3c0b2f81f5788681528dee8c532ad736535870d03","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a7ade9334433fe6af1e5b2d3c0b2f81f5788681528dee8c532ad736535870d03","first_computed_at":"2026-07-05T12:11:31.203985Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:11:31.203985Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"iNTzht8h0XlpF03fHA2k6cgkcazMxh+hZLyahDVkpT2C5TvLD/jONyfHi7Ve3BfltWS/PebeZVj++guYtleLBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T12:11:31.204519Z","signed_message":"canonical_sha256_bytes"},"source_id":"2509.10963","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a27f99814a1ff7d20e6b4f82829e137ae81509934137f63113f5a16e50448a5d","sha256:83aa9f726abbc5cb13c3a25c76e418b602a9e0da1ab21c23c70e8d4ab497644e"],"state_sha256":"23483a4ad1cdf9b61f278c33a4ae28bf340a1c86a6f5bdbd14f857103ec6e6cf"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kA8IJ4mx+86FLnTvvCGNbLFytRjV0AUPXreoWiAca5tUDjvo+QwukUziv9nEkRtIDoVDL9bEV+W4kxZEr/bGAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-27T02:26:56.998997Z","bundle_sha256":"6928d079e5626ca061fc91198061832ffc47ef2cd66bf6dc09660b61c086faf6"}}