{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:Z7OFH3JEQPVDTGRYCUL2LUCMNC","short_pith_number":"pith:Z7OFH3JE","canonical_record":{"source":{"id":"2505.18404","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-23T22:17:18Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"4430f9b7cdff639be9153fd7bd4b2dd763b37215000bcc0d256147860643cee0","abstract_canon_sha256":"fcf6d06b225c53d3f5b3995e44177189bd4a8fedfc048e682019426773411a91"},"schema_version":"1.0"},"canonical_sha256":"cfdc53ed2483ea399a381517a5d04c688a34d816c05eb29ac26fc59bdb6605c5","source":{"kind":"arxiv","id":"2505.18404","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.18404","created_at":"2026-07-05T11:08:45Z"},{"alias_kind":"arxiv_version","alias_value":"2505.18404v1","created_at":"2026-07-05T11:08:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.18404","created_at":"2026-07-05T11:08:45Z"},{"alias_kind":"pith_short_12","alias_value":"Z7OFH3JEQPVD","created_at":"2026-07-05T11:08:45Z"},{"alias_kind":"pith_short_16","alias_value":"Z7OFH3JEQPVDTGRY","created_at":"2026-07-05T11:08:45Z"},{"alias_kind":"pith_short_8","alias_value":"Z7OFH3JE","created_at":"2026-07-05T11:08:45Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:Z7OFH3JEQPVDTGRYCUL2LUCMNC","target":"record","payload":{"canonical_record":{"source":{"id":"2505.18404","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-23T22:17:18Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"4430f9b7cdff639be9153fd7bd4b2dd763b37215000bcc0d256147860643cee0","abstract_canon_sha256":"fcf6d06b225c53d3f5b3995e44177189bd4a8fedfc048e682019426773411a91"},"schema_version":"1.0"},"canonical_sha256":"cfdc53ed2483ea399a381517a5d04c688a34d816c05eb29ac26fc59bdb6605c5","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:08:45.906220Z","signature_b64":"Jj5xFNbTTPs2ekEt3seQ+aqFJNZkph6KE9M6YaIVmXAsPrFZ+k9dkV3+vblde12t71VmTdYJKlDRFZ0MZCCZCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cfdc53ed2483ea399a381517a5d04c688a34d816c05eb29ac26fc59bdb6605c5","last_reissued_at":"2026-07-05T11:08:45.905754Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:08:45.905754Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.18404","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":"2026-07-05T11:08:45Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NzKXPOyNDhCcHRui8wZzWoSWhRSqOQrPQ9n9RqwKCbFegRuXidIfYRUnvHQ+kscCiDnUr70p+7iI5xaH01KnAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T05:58:05.791680Z"},"content_sha256":"620622dda9ba423a6075608284cfbddf86db1f623e9744c91fbba37606b806fd","schema_version":"1.0","event_id":"sha256:620622dda9ba423a6075608284cfbddf86db1f623e9744c91fbba37606b806fd"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:Z7OFH3JEQPVDTGRYCUL2LUCMNC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Thought calibration: Efficient and confident test-time scaling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Cai Zhou, Menghua Wu, Stephen Bates, Tommi Jaakkola","submitted_at":"2025-05-23T22:17:18Z","abstract_excerpt":"Reasoning large language models achieve impressive test-time scaling by thinking for longer, but this performance gain comes at significant compute cost. Directly limiting test-time budget hurts overall performance, but not all problems are equally difficult. We propose thought calibration to decide dynamically when thinking can be terminated. To calibrate our decision rule, we view a language model's growing body of thoughts as a nested sequence of reasoning trees, where the goal is to identify the point at which novel reasoning plateaus. We realize this framework through lightweight probes t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.18404","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/2505.18404/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":"2026-07-05T11:08:45Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yOvo4gmjRS5q3eOB0Jg7hsPnWWh2B7WQXVxOYYHhfg5B0B1O03wwJaC8GiUnZCDFZCiRsPOKdmgSScIXy+kQAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T05:58:05.792348Z"},"content_sha256":"bb165c5aed0d3883a5cf1accad7aaac858178af9e503ae262951fe60622ffc7a","schema_version":"1.0","event_id":"sha256:bb165c5aed0d3883a5cf1accad7aaac858178af9e503ae262951fe60622ffc7a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/Z7OFH3JEQPVDTGRYCUL2LUCMNC/bundle.json","state_url":"https://pith.science/pith/Z7OFH3JEQPVDTGRYCUL2LUCMNC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/Z7OFH3JEQPVDTGRYCUL2LUCMNC/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":"2026-07-07T05:58:05Z","links":{"resolver":"https://pith.science/pith/Z7OFH3JEQPVDTGRYCUL2LUCMNC","bundle":"https://pith.science/pith/Z7OFH3JEQPVDTGRYCUL2LUCMNC/bundle.json","state":"https://pith.science/pith/Z7OFH3JEQPVDTGRYCUL2LUCMNC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/Z7OFH3JEQPVDTGRYCUL2LUCMNC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:Z7OFH3JEQPVDTGRYCUL2LUCMNC","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":"fcf6d06b225c53d3f5b3995e44177189bd4a8fedfc048e682019426773411a91","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-23T22:17:18Z","title_canon_sha256":"4430f9b7cdff639be9153fd7bd4b2dd763b37215000bcc0d256147860643cee0"},"schema_version":"1.0","source":{"id":"2505.18404","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.18404","created_at":"2026-07-05T11:08:45Z"},{"alias_kind":"arxiv_version","alias_value":"2505.18404v1","created_at":"2026-07-05T11:08:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.18404","created_at":"2026-07-05T11:08:45Z"},{"alias_kind":"pith_short_12","alias_value":"Z7OFH3JEQPVD","created_at":"2026-07-05T11:08:45Z"},{"alias_kind":"pith_short_16","alias_value":"Z7OFH3JEQPVDTGRY","created_at":"2026-07-05T11:08:45Z"},{"alias_kind":"pith_short_8","alias_value":"Z7OFH3JE","created_at":"2026-07-05T11:08:45Z"}],"graph_snapshots":[{"event_id":"sha256:bb165c5aed0d3883a5cf1accad7aaac858178af9e503ae262951fe60622ffc7a","target":"graph","created_at":"2026-07-05T11:08:45Z","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/2505.18404/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reasoning large language models achieve impressive test-time scaling by thinking for longer, but this performance gain comes at significant compute cost. Directly limiting test-time budget hurts overall performance, but not all problems are equally difficult. We propose thought calibration to decide dynamically when thinking can be terminated. To calibrate our decision rule, we view a language model's growing body of thoughts as a nested sequence of reasoning trees, where the goal is to identify the point at which novel reasoning plateaus. We realize this framework through lightweight probes t","authors_text":"Cai Zhou, Menghua Wu, Stephen Bates, Tommi Jaakkola","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-23T22:17:18Z","title":"Thought calibration: Efficient and confident test-time scaling"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.18404","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":"sha256:620622dda9ba423a6075608284cfbddf86db1f623e9744c91fbba37606b806fd","target":"record","created_at":"2026-07-05T11:08:45Z","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":"fcf6d06b225c53d3f5b3995e44177189bd4a8fedfc048e682019426773411a91","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-23T22:17:18Z","title_canon_sha256":"4430f9b7cdff639be9153fd7bd4b2dd763b37215000bcc0d256147860643cee0"},"schema_version":"1.0","source":{"id":"2505.18404","kind":"arxiv","version":1}},"canonical_sha256":"cfdc53ed2483ea399a381517a5d04c688a34d816c05eb29ac26fc59bdb6605c5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cfdc53ed2483ea399a381517a5d04c688a34d816c05eb29ac26fc59bdb6605c5","first_computed_at":"2026-07-05T11:08:45.905754Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:08:45.905754Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Jj5xFNbTTPs2ekEt3seQ+aqFJNZkph6KE9M6YaIVmXAsPrFZ+k9dkV3+vblde12t71VmTdYJKlDRFZ0MZCCZCA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:08:45.906220Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.18404","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:620622dda9ba423a6075608284cfbddf86db1f623e9744c91fbba37606b806fd","sha256:bb165c5aed0d3883a5cf1accad7aaac858178af9e503ae262951fe60622ffc7a"],"state_sha256":"8231849d87901bba9e8b68a337442ca82043f37f73941695fc5032c8612ea5e5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rDiBlE6cxEN6ACAw4PAf/fBUxxb7yXElytz3AivfOlWBOYRPTjm1bB6E4ZZfCVfe+7cKJNQchh7/xbrnkoSWCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T05:58:05.795556Z","bundle_sha256":"5c7528715189961675a6efd522f43e25fd533299757195bd85a86b4cd4646608"}}