{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:XCJWB4KQZYTNZ2JLGZIJKULSA7","short_pith_number":"pith:XCJWB4KQ","canonical_record":{"source":{"id":"2507.18122","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-24T06:17:39Z","cross_cats_sorted":[],"title_canon_sha256":"e9c5baf2025f14fa13a786ced0eb0f3f2f3984736286c16a799134ee470dca47","abstract_canon_sha256":"6c4600ccc9acd1e8315bd0927e3050fdc72f27800674eb78b81b3efbbcaf6713"},"schema_version":"1.0"},"canonical_sha256":"b89360f150ce26dce92b365095517207da2b207a6bbe43e974b907faed6bccf5","source":{"kind":"arxiv","id":"2507.18122","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.18122","created_at":"2026-07-05T11:42:42Z"},{"alias_kind":"arxiv_version","alias_value":"2507.18122v1","created_at":"2026-07-05T11:42:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.18122","created_at":"2026-07-05T11:42:42Z"},{"alias_kind":"pith_short_12","alias_value":"XCJWB4KQZYTN","created_at":"2026-07-05T11:42:42Z"},{"alias_kind":"pith_short_16","alias_value":"XCJWB4KQZYTNZ2JL","created_at":"2026-07-05T11:42:42Z"},{"alias_kind":"pith_short_8","alias_value":"XCJWB4KQ","created_at":"2026-07-05T11:42:42Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:XCJWB4KQZYTNZ2JLGZIJKULSA7","target":"record","payload":{"canonical_record":{"source":{"id":"2507.18122","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-24T06:17:39Z","cross_cats_sorted":[],"title_canon_sha256":"e9c5baf2025f14fa13a786ced0eb0f3f2f3984736286c16a799134ee470dca47","abstract_canon_sha256":"6c4600ccc9acd1e8315bd0927e3050fdc72f27800674eb78b81b3efbbcaf6713"},"schema_version":"1.0"},"canonical_sha256":"b89360f150ce26dce92b365095517207da2b207a6bbe43e974b907faed6bccf5","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:42:42.221708Z","signature_b64":"h0VNac3VSZZZGLwCkzeMfrDBuuChEt69dkspT4AeMjzhfxkL/OclTxhR/8QjFnGgPnqvYWzkUoEhQmB/JsKgCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b89360f150ce26dce92b365095517207da2b207a6bbe43e974b907faed6bccf5","last_reissued_at":"2026-07-05T11:42:42.221177Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:42:42.221177Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.18122","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":"2026-07-05T11:42:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"YRMZHAFra9T8zGfE8FfIT/6GPUwefIGkXnrgz6tkZ2CeLYPemFm9xk1b8DOadiFO5EiPuJ2vGqPegzS3x597AA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T07:07:06.905701Z"},"content_sha256":"29501a15c69b1956d9ce8a92d56ec19039a9d612b8d1a3cd89e492e67d7b293a","schema_version":"1.0","event_id":"sha256:29501a15c69b1956d9ce8a92d56ec19039a9d612b8d1a3cd89e492e67d7b293a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:XCJWB4KQZYTNZ2JLGZIJKULSA7","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Maximizing Prefix-Confidence at Test-Time Efficiently Improves Mathematical Reasoning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Andreas Krause, Ido Hakimi, Jonas H\\\"ubotter, Matthias Otth","submitted_at":"2025-07-24T06:17:39Z","abstract_excerpt":"Recent work has shown that language models can self-improve by maximizing their own confidence in their predictions, without relying on external verifiers or reward signals. In this work, we study the test-time scaling of language models for mathematical reasoning tasks, where the model's own confidence is used to select the most promising attempts. Surprisingly, we find that we can achieve significant performance gains by continuing only the most promising attempt, selected by the model's prefix-confidence. We systematically evaluate prefix-confidence scaling on five mathematical reasoning da"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.18122","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/2507.18122/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":"2026-07-05T11:42:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"McYs4AMEUX8SicBWfdWtmUncU/VI60SINPi5LwxiUXl2p6BRstGJ0MiJCTj74MqsJb/KsSM1HGuzioBCiVepDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T07:07:06.906654Z"},"content_sha256":"75415e595ce4f01969b36913c40962fed2c3dbc95b3a7982613363a99f0879b9","schema_version":"1.0","event_id":"sha256:75415e595ce4f01969b36913c40962fed2c3dbc95b3a7982613363a99f0879b9"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XCJWB4KQZYTNZ2JLGZIJKULSA7/bundle.json","state_url":"https://pith.science/pith/XCJWB4KQZYTNZ2JLGZIJKULSA7/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XCJWB4KQZYTNZ2JLGZIJKULSA7/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":"2026-08-14T07:07:06Z","links":{"resolver":"https://pith.science/pith/XCJWB4KQZYTNZ2JLGZIJKULSA7","bundle":"https://pith.science/pith/XCJWB4KQZYTNZ2JLGZIJKULSA7/bundle.json","state":"https://pith.science/pith/XCJWB4KQZYTNZ2JLGZIJKULSA7/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XCJWB4KQZYTNZ2JLGZIJKULSA7/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:XCJWB4KQZYTNZ2JLGZIJKULSA7","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":"6c4600ccc9acd1e8315bd0927e3050fdc72f27800674eb78b81b3efbbcaf6713","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-24T06:17:39Z","title_canon_sha256":"e9c5baf2025f14fa13a786ced0eb0f3f2f3984736286c16a799134ee470dca47"},"schema_version":"1.0","source":{"id":"2507.18122","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.18122","created_at":"2026-07-05T11:42:42Z"},{"alias_kind":"arxiv_version","alias_value":"2507.18122v1","created_at":"2026-07-05T11:42:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.18122","created_at":"2026-07-05T11:42:42Z"},{"alias_kind":"pith_short_12","alias_value":"XCJWB4KQZYTN","created_at":"2026-07-05T11:42:42Z"},{"alias_kind":"pith_short_16","alias_value":"XCJWB4KQZYTNZ2JL","created_at":"2026-07-05T11:42:42Z"},{"alias_kind":"pith_short_8","alias_value":"XCJWB4KQ","created_at":"2026-07-05T11:42:42Z"}],"graph_snapshots":[{"event_id":"sha256:75415e595ce4f01969b36913c40962fed2c3dbc95b3a7982613363a99f0879b9","target":"graph","created_at":"2026-07-05T11:42:42Z","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/2507.18122/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent work has shown that language models can self-improve by maximizing their own confidence in their predictions, without relying on external verifiers or reward signals. In this work, we study the test-time scaling of language models for mathematical reasoning tasks, where the model's own confidence is used to select the most promising attempts. Surprisingly, we find that we can achieve significant performance gains by continuing only the most promising attempt, selected by the model's prefix-confidence. We systematically evaluate prefix-confidence scaling on five mathematical reasoning da","authors_text":"Andreas Krause, Ido Hakimi, Jonas H\\\"ubotter, Matthias Otth","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-24T06:17:39Z","title":"Maximizing Prefix-Confidence at Test-Time Efficiently Improves Mathematical Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.18122","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":"sha256:29501a15c69b1956d9ce8a92d56ec19039a9d612b8d1a3cd89e492e67d7b293a","target":"record","created_at":"2026-07-05T11:42:42Z","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":"6c4600ccc9acd1e8315bd0927e3050fdc72f27800674eb78b81b3efbbcaf6713","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-24T06:17:39Z","title_canon_sha256":"e9c5baf2025f14fa13a786ced0eb0f3f2f3984736286c16a799134ee470dca47"},"schema_version":"1.0","source":{"id":"2507.18122","kind":"arxiv","version":1}},"canonical_sha256":"b89360f150ce26dce92b365095517207da2b207a6bbe43e974b907faed6bccf5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b89360f150ce26dce92b365095517207da2b207a6bbe43e974b907faed6bccf5","first_computed_at":"2026-07-05T11:42:42.221177Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:42:42.221177Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"h0VNac3VSZZZGLwCkzeMfrDBuuChEt69dkspT4AeMjzhfxkL/OclTxhR/8QjFnGgPnqvYWzkUoEhQmB/JsKgCA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:42:42.221708Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.18122","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:29501a15c69b1956d9ce8a92d56ec19039a9d612b8d1a3cd89e492e67d7b293a","sha256:75415e595ce4f01969b36913c40962fed2c3dbc95b3a7982613363a99f0879b9"],"state_sha256":"7bfeea374223f926ef4148ea8264ad41c7ffcf967bfc44e333fd3750427dd5ba"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Bw4GxH4myb9iqe985tUnhjryTR4p0DrCtXWU1qhEevZ61ConnWN8lvGSZWpW2dzcHAL37wwxVd8Cgxy+IqLAAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-14T07:07:06.913735Z","bundle_sha256":"a029ff666a3b782601b5de1addc455534044b2638c16342035eb0bf1c67d95cf"}}