{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:ULDMHZU7U5T4UTSER3XROQL55A","short_pith_number":"pith:ULDMHZU7","canonical_record":{"source":{"id":"2601.13735","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-01-20T08:46:33Z","cross_cats_sorted":[],"title_canon_sha256":"7b39d097ef05948346c6dce12c042c530edb75e14177342b13f318d935cb1649","abstract_canon_sha256":"5773d29aa6acfae29cf5ca25d7676f4868581b65f5435efb4caeb26fd7e0a5c1"},"schema_version":"1.0"},"canonical_sha256":"a2c6c3e69fa767ca4e448eef17417de8127305c1000497785aa7f66142cd7479","source":{"kind":"arxiv","id":"2601.13735","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2601.13735","created_at":"2026-06-04T01:08:40Z"},{"alias_kind":"arxiv_version","alias_value":"2601.13735v2","created_at":"2026-06-04T01:08:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.13735","created_at":"2026-06-04T01:08:40Z"},{"alias_kind":"pith_short_12","alias_value":"ULDMHZU7U5T4","created_at":"2026-06-04T01:08:40Z"},{"alias_kind":"pith_short_16","alias_value":"ULDMHZU7U5T4UTSE","created_at":"2026-06-04T01:08:40Z"},{"alias_kind":"pith_short_8","alias_value":"ULDMHZU7","created_at":"2026-06-04T01:08:40Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:ULDMHZU7U5T4UTSER3XROQL55A","target":"record","payload":{"canonical_record":{"source":{"id":"2601.13735","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-01-20T08:46:33Z","cross_cats_sorted":[],"title_canon_sha256":"7b39d097ef05948346c6dce12c042c530edb75e14177342b13f318d935cb1649","abstract_canon_sha256":"5773d29aa6acfae29cf5ca25d7676f4868581b65f5435efb4caeb26fd7e0a5c1"},"schema_version":"1.0"},"canonical_sha256":"a2c6c3e69fa767ca4e448eef17417de8127305c1000497785aa7f66142cd7479","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-04T01:08:40.173397Z","signature_b64":"RjjmuLuqCLVsQQKzhvfU/MvaWsUrz9Uo6iX6gQCRw9DfJWkiex56tgG2s2yK3id/br0IQYgT0MHPH+X8nr4VDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a2c6c3e69fa767ca4e448eef17417de8127305c1000497785aa7f66142cd7479","last_reissued_at":"2026-06-04T01:08:40.172499Z","signature_status":"signed_v1","first_computed_at":"2026-06-04T01:08:40.172499Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2601.13735","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-04T01:08:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"i9ggzhXwirjnbAKHP/7U/BQIZMFF3b40XgYy+pNxUpKc92nKXgxkU9I0K3Nb2u3KfpigQ4+WqyAc0boRu8QyCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-11T00:51:47.539927Z"},"content_sha256":"5844c506babd3e92a5158a20268b31d4e23c7344de8954d142a8bf1884509c01","schema_version":"1.0","event_id":"sha256:5844c506babd3e92a5158a20268b31d4e23c7344de8954d142a8bf1884509c01"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:ULDMHZU7U5T4UTSER3XROQL55A","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Reasoning or Fluency? Dissecting Probabilistic Confidence in Best-of-N Selection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Hojin Kim, Jaehyung Kim","submitted_at":"2026-01-20T08:46:33Z","abstract_excerpt":"Probabilistic confidence metrics are increasingly adopted as proxies for reasoning quality in Best-of-N selection, under the assumption that higher confidence reflects higher reasoning fidelity. In this work, we challenge this assumption by investigating whether these metrics truly capture inter-step causal dependencies necessary for valid reasoning. We introduce three classes of inter-step causality perturbations that systematically disrupt dependencies between reasoning steps while preserving local fluency. Surprisingly, across diverse model families and reasoning benchmarks, we find that se"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.13735","kind":"arxiv","version":2},"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/2601.13735/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":"2026-06-04T01:08:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"QvvI8raKY6aS8ydt4jM+/R1mmHPwaH0RpqCIVB87ywXbrC4+qMUCDMTTTj2P3fuowoXeMdGCBc0LhoWJyXNlDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-11T00:51:47.540730Z"},"content_sha256":"cc81a406d052db16070ca6419bc9650b25ee76f0610d110b636ef47705876c02","schema_version":"1.0","event_id":"sha256:cc81a406d052db16070ca6419bc9650b25ee76f0610d110b636ef47705876c02"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ULDMHZU7U5T4UTSER3XROQL55A/bundle.json","state_url":"https://pith.science/pith/ULDMHZU7U5T4UTSER3XROQL55A/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ULDMHZU7U5T4UTSER3XROQL55A/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":"2026-06-11T00:51:47Z","links":{"resolver":"https://pith.science/pith/ULDMHZU7U5T4UTSER3XROQL55A","bundle":"https://pith.science/pith/ULDMHZU7U5T4UTSER3XROQL55A/bundle.json","state":"https://pith.science/pith/ULDMHZU7U5T4UTSER3XROQL55A/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ULDMHZU7U5T4UTSER3XROQL55A/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:ULDMHZU7U5T4UTSER3XROQL55A","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":"5773d29aa6acfae29cf5ca25d7676f4868581b65f5435efb4caeb26fd7e0a5c1","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-01-20T08:46:33Z","title_canon_sha256":"7b39d097ef05948346c6dce12c042c530edb75e14177342b13f318d935cb1649"},"schema_version":"1.0","source":{"id":"2601.13735","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2601.13735","created_at":"2026-06-04T01:08:40Z"},{"alias_kind":"arxiv_version","alias_value":"2601.13735v2","created_at":"2026-06-04T01:08:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.13735","created_at":"2026-06-04T01:08:40Z"},{"alias_kind":"pith_short_12","alias_value":"ULDMHZU7U5T4","created_at":"2026-06-04T01:08:40Z"},{"alias_kind":"pith_short_16","alias_value":"ULDMHZU7U5T4UTSE","created_at":"2026-06-04T01:08:40Z"},{"alias_kind":"pith_short_8","alias_value":"ULDMHZU7","created_at":"2026-06-04T01:08:40Z"}],"graph_snapshots":[{"event_id":"sha256:cc81a406d052db16070ca6419bc9650b25ee76f0610d110b636ef47705876c02","target":"graph","created_at":"2026-06-04T01:08:40Z","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/2601.13735/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Probabilistic confidence metrics are increasingly adopted as proxies for reasoning quality in Best-of-N selection, under the assumption that higher confidence reflects higher reasoning fidelity. In this work, we challenge this assumption by investigating whether these metrics truly capture inter-step causal dependencies necessary for valid reasoning. We introduce three classes of inter-step causality perturbations that systematically disrupt dependencies between reasoning steps while preserving local fluency. Surprisingly, across diverse model families and reasoning benchmarks, we find that se","authors_text":"Hojin Kim, Jaehyung Kim","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-01-20T08:46:33Z","title":"Reasoning or Fluency? Dissecting Probabilistic Confidence in Best-of-N Selection"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.13735","kind":"arxiv","version":2},"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":"sha256:5844c506babd3e92a5158a20268b31d4e23c7344de8954d142a8bf1884509c01","target":"record","created_at":"2026-06-04T01:08:40Z","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":"5773d29aa6acfae29cf5ca25d7676f4868581b65f5435efb4caeb26fd7e0a5c1","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-01-20T08:46:33Z","title_canon_sha256":"7b39d097ef05948346c6dce12c042c530edb75e14177342b13f318d935cb1649"},"schema_version":"1.0","source":{"id":"2601.13735","kind":"arxiv","version":2}},"canonical_sha256":"a2c6c3e69fa767ca4e448eef17417de8127305c1000497785aa7f66142cd7479","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a2c6c3e69fa767ca4e448eef17417de8127305c1000497785aa7f66142cd7479","first_computed_at":"2026-06-04T01:08:40.172499Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-04T01:08:40.172499Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"RjjmuLuqCLVsQQKzhvfU/MvaWsUrz9Uo6iX6gQCRw9DfJWkiex56tgG2s2yK3id/br0IQYgT0MHPH+X8nr4VDw==","signature_status":"signed_v1","signed_at":"2026-06-04T01:08:40.173397Z","signed_message":"canonical_sha256_bytes"},"source_id":"2601.13735","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5844c506babd3e92a5158a20268b31d4e23c7344de8954d142a8bf1884509c01","sha256:cc81a406d052db16070ca6419bc9650b25ee76f0610d110b636ef47705876c02"],"state_sha256":"9f8e80c3adb083a9edd53902301789f16a5dccb9c6e91954e52c7901aa1447b7"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"I/gTflFvTA6qP0yGcXOmfZEn6we/HIyj5gNIrInnSPYWYGZPaZDXxZslU9TQeH3T6zDprID+NN46Jv13UOCaDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-11T00:51:47.545334Z","bundle_sha256":"00cb8d3dfd3f31842f3ffd05f8fad4a66f1970229f2972eea1885caebafa938a"}}