{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:LGS4NBXUUDOBRCQZLPABABMGM7","short_pith_number":"pith:LGS4NBXU","canonical_record":{"source":{"id":"2506.07947","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-09T17:11:07Z","cross_cats_sorted":[],"title_canon_sha256":"f90d567640153bfc860528a42d7b382897219e485b42931874d10d267da44df3","abstract_canon_sha256":"02bda4200cd7920ba5ee94853b99ed818fb90f6accb0f344e5de5341635efde7"},"schema_version":"1.0"},"canonical_sha256":"59a5c686f4a0dc188a195bc010058667f149d6ac4b47364544196eea618cc39f","source":{"kind":"arxiv","id":"2506.07947","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.07947","created_at":"2026-07-05T11:18:35Z"},{"alias_kind":"arxiv_version","alias_value":"2506.07947v1","created_at":"2026-07-05T11:18:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.07947","created_at":"2026-07-05T11:18:35Z"},{"alias_kind":"pith_short_12","alias_value":"LGS4NBXUUDOB","created_at":"2026-07-05T11:18:35Z"},{"alias_kind":"pith_short_16","alias_value":"LGS4NBXUUDOBRCQZ","created_at":"2026-07-05T11:18:35Z"},{"alias_kind":"pith_short_8","alias_value":"LGS4NBXU","created_at":"2026-07-05T11:18:35Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:LGS4NBXUUDOBRCQZLPABABMGM7","target":"record","payload":{"canonical_record":{"source":{"id":"2506.07947","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-09T17:11:07Z","cross_cats_sorted":[],"title_canon_sha256":"f90d567640153bfc860528a42d7b382897219e485b42931874d10d267da44df3","abstract_canon_sha256":"02bda4200cd7920ba5ee94853b99ed818fb90f6accb0f344e5de5341635efde7"},"schema_version":"1.0"},"canonical_sha256":"59a5c686f4a0dc188a195bc010058667f149d6ac4b47364544196eea618cc39f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:18:35.873481Z","signature_b64":"mzU5XSZmQG5yOEcv/ZB9X02FqvGKL5iZIQQa0Za5fnGWj0d9LTTSyp/r529ZS71pd//v/npeYr+Vf035+Bk5AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"59a5c686f4a0dc188a195bc010058667f149d6ac4b47364544196eea618cc39f","last_reissued_at":"2026-07-05T11:18:35.872925Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:18:35.872925Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.07947","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":"2026-07-05T11:18:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hohfE39x15WL22lFrwRpS+uDfUTBthzDHJFGKvG58jt0oXuQJUVmayg3DTh/gfnYyDYTtFESm3r0HGQZI8xKCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T15:43:57.750604Z"},"content_sha256":"ba82ed4194b87584f8df97dbc335911f2b015b5be5efdcf0ca7995f967637a77","schema_version":"1.0","event_id":"sha256:ba82ed4194b87584f8df97dbc335911f2b015b5be5efdcf0ca7995f967637a77"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:LGS4NBXUUDOBRCQZLPABABMGM7","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Statistical Hypothesis Testing for Auditing Robustness in Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Mihaela van der Schaar, Paulius Rauba, Qiyao Wei","submitted_at":"2025-06-09T17:11:07Z","abstract_excerpt":"Consider the problem of testing whether the outputs of a large language model (LLM) system change under an arbitrary intervention, such as an input perturbation or changing the model variant. We cannot simply compare two LLM outputs since they might differ due to the stochastic nature of the system, nor can we compare the entire output distribution due to computational intractability. While existing methods for analyzing text-based outputs exist, they focus on fundamentally different problems, such as measuring bias or fairness. To this end, we introduce distribution-based perturbation analysi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.07947","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/2506.07947/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":"2026-07-05T11:18:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"W74RmypV8FaR9CDDl+OC+js57PwlZ1pIt2gwKknw95k4AXgou1SFO2Sru9AdUc06byNrZJJOHIZNzjpOY5hlBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T15:43:57.751178Z"},"content_sha256":"717e4cab09986f63c8075164cfc2eba465ef2ad48697313095a35219e84f2669","schema_version":"1.0","event_id":"sha256:717e4cab09986f63c8075164cfc2eba465ef2ad48697313095a35219e84f2669"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LGS4NBXUUDOBRCQZLPABABMGM7/bundle.json","state_url":"https://pith.science/pith/LGS4NBXUUDOBRCQZLPABABMGM7/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LGS4NBXUUDOBRCQZLPABABMGM7/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":"2026-08-09T15:43:57Z","links":{"resolver":"https://pith.science/pith/LGS4NBXUUDOBRCQZLPABABMGM7","bundle":"https://pith.science/pith/LGS4NBXUUDOBRCQZLPABABMGM7/bundle.json","state":"https://pith.science/pith/LGS4NBXUUDOBRCQZLPABABMGM7/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LGS4NBXUUDOBRCQZLPABABMGM7/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:LGS4NBXUUDOBRCQZLPABABMGM7","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":"02bda4200cd7920ba5ee94853b99ed818fb90f6accb0f344e5de5341635efde7","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-09T17:11:07Z","title_canon_sha256":"f90d567640153bfc860528a42d7b382897219e485b42931874d10d267da44df3"},"schema_version":"1.0","source":{"id":"2506.07947","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.07947","created_at":"2026-07-05T11:18:35Z"},{"alias_kind":"arxiv_version","alias_value":"2506.07947v1","created_at":"2026-07-05T11:18:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.07947","created_at":"2026-07-05T11:18:35Z"},{"alias_kind":"pith_short_12","alias_value":"LGS4NBXUUDOB","created_at":"2026-07-05T11:18:35Z"},{"alias_kind":"pith_short_16","alias_value":"LGS4NBXUUDOBRCQZ","created_at":"2026-07-05T11:18:35Z"},{"alias_kind":"pith_short_8","alias_value":"LGS4NBXU","created_at":"2026-07-05T11:18:35Z"}],"graph_snapshots":[{"event_id":"sha256:717e4cab09986f63c8075164cfc2eba465ef2ad48697313095a35219e84f2669","target":"graph","created_at":"2026-07-05T11:18:35Z","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/2506.07947/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Consider the problem of testing whether the outputs of a large language model (LLM) system change under an arbitrary intervention, such as an input perturbation or changing the model variant. We cannot simply compare two LLM outputs since they might differ due to the stochastic nature of the system, nor can we compare the entire output distribution due to computational intractability. While existing methods for analyzing text-based outputs exist, they focus on fundamentally different problems, such as measuring bias or fairness. To this end, we introduce distribution-based perturbation analysi","authors_text":"Mihaela van der Schaar, Paulius Rauba, Qiyao Wei","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-09T17:11:07Z","title":"Statistical Hypothesis Testing for Auditing Robustness in Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.07947","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:ba82ed4194b87584f8df97dbc335911f2b015b5be5efdcf0ca7995f967637a77","target":"record","created_at":"2026-07-05T11:18:35Z","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":"02bda4200cd7920ba5ee94853b99ed818fb90f6accb0f344e5de5341635efde7","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-09T17:11:07Z","title_canon_sha256":"f90d567640153bfc860528a42d7b382897219e485b42931874d10d267da44df3"},"schema_version":"1.0","source":{"id":"2506.07947","kind":"arxiv","version":1}},"canonical_sha256":"59a5c686f4a0dc188a195bc010058667f149d6ac4b47364544196eea618cc39f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"59a5c686f4a0dc188a195bc010058667f149d6ac4b47364544196eea618cc39f","first_computed_at":"2026-07-05T11:18:35.872925Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:18:35.872925Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"mzU5XSZmQG5yOEcv/ZB9X02FqvGKL5iZIQQa0Za5fnGWj0d9LTTSyp/r529ZS71pd//v/npeYr+Vf035+Bk5AA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:18:35.873481Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.07947","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ba82ed4194b87584f8df97dbc335911f2b015b5be5efdcf0ca7995f967637a77","sha256:717e4cab09986f63c8075164cfc2eba465ef2ad48697313095a35219e84f2669"],"state_sha256":"9cfb015fc3ca7b7da78dbf19951954d132fd8a67e3eeca35276aba3d29873737"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rBetKJws7zKyAS73rT5JdcbH95x6Q9y0SQ0ovtlO7WC/xIk7Nht1/avW+p5n69YkF7I5T1nKQMTg1fURf7srAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T15:43:57.756413Z","bundle_sha256":"29962449cc94715d98c93c4d660aba2b8ed8633473af2c38d0f0f145e3bdde94"}}