{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:P5LKYUFP32GQGNDM4EGVOIMPSG","short_pith_number":"pith:P5LKYUFP","canonical_record":{"source":{"id":"2509.09396","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-11T12:25:41Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"82f7fb2c530dd0d29bc0f81f0933ff2039d4094084a101358f0a60b029822fee","abstract_canon_sha256":"c20349bd2928490693132d38504316d05d714d6acc5822cff771dec263a20617"},"schema_version":"1.0"},"canonical_sha256":"7f56ac50afde8d03346ce10d57218f9188ac9bd057e502492d0ea271e1796cb1","source":{"kind":"arxiv","id":"2509.09396","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.09396","created_at":"2026-07-05T12:09:35Z"},{"alias_kind":"arxiv_version","alias_value":"2509.09396v1","created_at":"2026-07-05T12:09:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.09396","created_at":"2026-07-05T12:09:35Z"},{"alias_kind":"pith_short_12","alias_value":"P5LKYUFP32GQ","created_at":"2026-07-05T12:09:35Z"},{"alias_kind":"pith_short_16","alias_value":"P5LKYUFP32GQGNDM","created_at":"2026-07-05T12:09:35Z"},{"alias_kind":"pith_short_8","alias_value":"P5LKYUFP","created_at":"2026-07-05T12:09:35Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:P5LKYUFP32GQGNDM4EGVOIMPSG","target":"record","payload":{"canonical_record":{"source":{"id":"2509.09396","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-11T12:25:41Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"82f7fb2c530dd0d29bc0f81f0933ff2039d4094084a101358f0a60b029822fee","abstract_canon_sha256":"c20349bd2928490693132d38504316d05d714d6acc5822cff771dec263a20617"},"schema_version":"1.0"},"canonical_sha256":"7f56ac50afde8d03346ce10d57218f9188ac9bd057e502492d0ea271e1796cb1","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:09:35.419978Z","signature_b64":"J5H3gGm211EsYeEIcgA3ctJZVSccSZhVQgeKF4GO2QYx5XJzkpeC7mTGj/9t+zVQOvq/nlyep/tuAW/IElXQAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7f56ac50afde8d03346ce10d57218f9188ac9bd057e502492d0ea271e1796cb1","last_reissued_at":"2026-07-05T12:09:35.419429Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:09:35.419429Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2509.09396","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":"2026-07-05T12:09:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gMX1as68iu60yTioOhpQiPVog/jIvucKhb3u939e7ULG6KCAkCSltnqpr5u+LvQMgY2iE9h4EYPgH/d1hhZqAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T13:21:24.302975Z"},"content_sha256":"220c260fcd4fee49656435db8a5496c27f3db67e9c82113e3bab1a920037ba8e","schema_version":"1.0","event_id":"sha256:220c260fcd4fee49656435db8a5496c27f3db67e9c82113e3bab1a920037ba8e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:P5LKYUFP32GQGNDM4EGVOIMPSG","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"LLMs Don't Know Their Own Decision Boundaries: The Unreliability of Self-Generated Counterfactual Explanations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Adam Mahdi, Andrew M. Bean, Chris Russell, Eoin Delaney, Harry Mayne, Ryan Othniel Kearns, Yushi Yang","submitted_at":"2025-09-11T12:25:41Z","abstract_excerpt":"To collaborate effectively with humans, language models must be able to explain their decisions in natural language. We study a specific type of self-explanation: self-generated counterfactual explanations (SCEs), where a model explains its prediction by modifying the input such that it would have predicted a different outcome. We evaluate whether LLMs can produce SCEs that are valid, achieving the intended outcome, and minimal, modifying the input no more than necessary. When asked to generate counterfactuals, we find that LLMs typically produce SCEs that are valid, but far from minimal, offe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.09396","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/2509.09396/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":"2026-07-05T12:09:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"bvGON5TXlX/jWRvAkXGsVrTvhAckvzmIAhdajGWrOrnbsbuycju+juKqeE8JUXZnEeJwsIx1PAu9jRHemD6FBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T13:21:24.303906Z"},"content_sha256":"5e8cde40117c533bc54f326b3954503bdc7af517cc1bb5c4ada71468b0b78343","schema_version":"1.0","event_id":"sha256:5e8cde40117c533bc54f326b3954503bdc7af517cc1bb5c4ada71468b0b78343"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/P5LKYUFP32GQGNDM4EGVOIMPSG/bundle.json","state_url":"https://pith.science/pith/P5LKYUFP32GQGNDM4EGVOIMPSG/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/P5LKYUFP32GQGNDM4EGVOIMPSG/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":"2026-08-05T13:21:24Z","links":{"resolver":"https://pith.science/pith/P5LKYUFP32GQGNDM4EGVOIMPSG","bundle":"https://pith.science/pith/P5LKYUFP32GQGNDM4EGVOIMPSG/bundle.json","state":"https://pith.science/pith/P5LKYUFP32GQGNDM4EGVOIMPSG/state.json","well_known_bundle":"https://pith.science/.well-known/pith/P5LKYUFP32GQGNDM4EGVOIMPSG/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:P5LKYUFP32GQGNDM4EGVOIMPSG","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":"c20349bd2928490693132d38504316d05d714d6acc5822cff771dec263a20617","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-11T12:25:41Z","title_canon_sha256":"82f7fb2c530dd0d29bc0f81f0933ff2039d4094084a101358f0a60b029822fee"},"schema_version":"1.0","source":{"id":"2509.09396","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.09396","created_at":"2026-07-05T12:09:35Z"},{"alias_kind":"arxiv_version","alias_value":"2509.09396v1","created_at":"2026-07-05T12:09:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.09396","created_at":"2026-07-05T12:09:35Z"},{"alias_kind":"pith_short_12","alias_value":"P5LKYUFP32GQ","created_at":"2026-07-05T12:09:35Z"},{"alias_kind":"pith_short_16","alias_value":"P5LKYUFP32GQGNDM","created_at":"2026-07-05T12:09:35Z"},{"alias_kind":"pith_short_8","alias_value":"P5LKYUFP","created_at":"2026-07-05T12:09:35Z"}],"graph_snapshots":[{"event_id":"sha256:5e8cde40117c533bc54f326b3954503bdc7af517cc1bb5c4ada71468b0b78343","target":"graph","created_at":"2026-07-05T12:09:35Z","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/2509.09396/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"To collaborate effectively with humans, language models must be able to explain their decisions in natural language. We study a specific type of self-explanation: self-generated counterfactual explanations (SCEs), where a model explains its prediction by modifying the input such that it would have predicted a different outcome. We evaluate whether LLMs can produce SCEs that are valid, achieving the intended outcome, and minimal, modifying the input no more than necessary. When asked to generate counterfactuals, we find that LLMs typically produce SCEs that are valid, but far from minimal, offe","authors_text":"Adam Mahdi, Andrew M. Bean, Chris Russell, Eoin Delaney, Harry Mayne, Ryan Othniel Kearns, Yushi Yang","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-11T12:25:41Z","title":"LLMs Don't Know Their Own Decision Boundaries: The Unreliability of Self-Generated Counterfactual Explanations"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.09396","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":"sha256:220c260fcd4fee49656435db8a5496c27f3db67e9c82113e3bab1a920037ba8e","target":"record","created_at":"2026-07-05T12:09:35Z","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":"c20349bd2928490693132d38504316d05d714d6acc5822cff771dec263a20617","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-11T12:25:41Z","title_canon_sha256":"82f7fb2c530dd0d29bc0f81f0933ff2039d4094084a101358f0a60b029822fee"},"schema_version":"1.0","source":{"id":"2509.09396","kind":"arxiv","version":1}},"canonical_sha256":"7f56ac50afde8d03346ce10d57218f9188ac9bd057e502492d0ea271e1796cb1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7f56ac50afde8d03346ce10d57218f9188ac9bd057e502492d0ea271e1796cb1","first_computed_at":"2026-07-05T12:09:35.419429Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:09:35.419429Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"J5H3gGm211EsYeEIcgA3ctJZVSccSZhVQgeKF4GO2QYx5XJzkpeC7mTGj/9t+zVQOvq/nlyep/tuAW/IElXQAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T12:09:35.419978Z","signed_message":"canonical_sha256_bytes"},"source_id":"2509.09396","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:220c260fcd4fee49656435db8a5496c27f3db67e9c82113e3bab1a920037ba8e","sha256:5e8cde40117c533bc54f326b3954503bdc7af517cc1bb5c4ada71468b0b78343"],"state_sha256":"9cb0b558e107e9368d3734263df7ef80f980212aebe3b97515a988a01b8958d6"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uNvBOjWkIgEaet/bbkGgSpOb+7/dq/LRYUo0HEvY+42Ys2NJAArbrqxlKksC9woghbtfKwZFinw3TOt8O1lNCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T13:21:24.309661Z","bundle_sha256":"50ac55d15a834699f46d308c1817758418cd3c5ba3cc80adf4756a92bd2ab6e2"}}