{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:3N35E2S3P6XIHBDHDZGRKUBNH6","short_pith_number":"pith:3N35E2S3","schema_version":"1.0","canonical_sha256":"db77d26a5b7fae8384671e4d15502d3f955219abed3776cac661158cf5c49748","source":{"kind":"arxiv","id":"2607.01612","version":1},"attestation_state":"computed","paper":{"title":"Scaling with Confidence: Calibrating Confidence of LLMs for Adaptive Test Time Scaling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Shanzhe Lei, Xuhong Wang, Xuqing Yang, Yi Yuan","submitted_at":"2026-07-02T02:29:33Z","abstract_excerpt":"Training large language models (LLMs) with reinforcement learning (RL) has significantly advanced their performance on reasoning and question-answering tasks. However, prevailing RL reward designs typically prioritize response correctness, neglecting to incentivize models to express their confidence accurately. This leads to a critical problem: performance gains are often accompanied by poor calibration between confidence and accuracy, misleading models to overconfidently hallucinate when uncertain. To address this limitation, we propose $\\textbf{C}$orrectness and $\\textbf{C}$onfidence $\\textb"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2607.01612","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-02T02:29:33Z","cross_cats_sorted":[],"title_canon_sha256":"75c8bc537aa8cf6b9f2adf1ba8baa3200d4dac2d62d62cb6e16c3659c5b177cb","abstract_canon_sha256":"7891a59388006cad26103417b4e4b0a97274c1343ae7fae51b9c19c28fc36d78"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-03T01:17:03.345117Z","signature_b64":"j57eYL7gd8xb4pCjkTOfUr0CiESImXg1sdT/TLtEFyM4DngMoH1UvsbVb3FfHtVhdSkYjGv1WZKj6YFwClGsBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"db77d26a5b7fae8384671e4d15502d3f955219abed3776cac661158cf5c49748","last_reissued_at":"2026-07-03T01:17:03.344721Z","signature_status":"signed_v1","first_computed_at":"2026-07-03T01:17:03.344721Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Scaling with Confidence: Calibrating Confidence of LLMs for Adaptive Test Time Scaling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Shanzhe Lei, Xuhong Wang, Xuqing Yang, Yi Yuan","submitted_at":"2026-07-02T02:29:33Z","abstract_excerpt":"Training large language models (LLMs) with reinforcement learning (RL) has significantly advanced their performance on reasoning and question-answering tasks. However, prevailing RL reward designs typically prioritize response correctness, neglecting to incentivize models to express their confidence accurately. This leads to a critical problem: performance gains are often accompanied by poor calibration between confidence and accuracy, misleading models to overconfidently hallucinate when uncertain. To address this limitation, we propose $\\textbf{C}$orrectness and $\\textbf{C}$onfidence $\\textb"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.01612","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/2607.01612/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2607.01612","created_at":"2026-07-03T01:17:03.344769+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.01612v1","created_at":"2026-07-03T01:17:03.344769+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.01612","created_at":"2026-07-03T01:17:03.344769+00:00"},{"alias_kind":"pith_short_12","alias_value":"3N35E2S3P6XI","created_at":"2026-07-03T01:17:03.344769+00:00"},{"alias_kind":"pith_short_16","alias_value":"3N35E2S3P6XIHBDH","created_at":"2026-07-03T01:17:03.344769+00:00"},{"alias_kind":"pith_short_8","alias_value":"3N35E2S3","created_at":"2026-07-03T01:17:03.344769+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3N35E2S3P6XIHBDHDZGRKUBNH6","json":"https://pith.science/pith/3N35E2S3P6XIHBDHDZGRKUBNH6.json","graph_json":"https://pith.science/api/pith-number/3N35E2S3P6XIHBDHDZGRKUBNH6/graph.json","events_json":"https://pith.science/api/pith-number/3N35E2S3P6XIHBDHDZGRKUBNH6/events.json","paper":"https://pith.science/paper/3N35E2S3"},"agent_actions":{"view_html":"https://pith.science/pith/3N35E2S3P6XIHBDHDZGRKUBNH6","download_json":"https://pith.science/pith/3N35E2S3P6XIHBDHDZGRKUBNH6.json","view_paper":"https://pith.science/paper/3N35E2S3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.01612&json=true","fetch_graph":"https://pith.science/api/pith-number/3N35E2S3P6XIHBDHDZGRKUBNH6/graph.json","fetch_events":"https://pith.science/api/pith-number/3N35E2S3P6XIHBDHDZGRKUBNH6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3N35E2S3P6XIHBDHDZGRKUBNH6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3N35E2S3P6XIHBDHDZGRKUBNH6/action/storage_attestation","attest_author":"https://pith.science/pith/3N35E2S3P6XIHBDHDZGRKUBNH6/action/author_attestation","sign_citation":"https://pith.science/pith/3N35E2S3P6XIHBDHDZGRKUBNH6/action/citation_signature","submit_replication":"https://pith.science/pith/3N35E2S3P6XIHBDHDZGRKUBNH6/action/replication_record"}},"created_at":"2026-07-03T01:17:03.344769+00:00","updated_at":"2026-07-03T01:17:03.344769+00:00"}