{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:BPQ4QZLWB5WICIN7UEBTBVWOTK","short_pith_number":"pith:BPQ4QZLW","schema_version":"1.0","canonical_sha256":"0be1c865760f6c8121bfa10330d6ce9abad7b1e0bd620e6bd0e612eb82f7c454","source":{"kind":"arxiv","id":"1706.04599","version":2},"attestation_state":"computed","paper":{"title":"On Calibration of Modern Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Chuan Guo, Geoff Pleiss, Kilian Q. Weinberger, Yu Sun","submitted_at":"2017-06-14T17:33:50Z","abstract_excerpt":"Confidence calibration -- the problem of predicting probability estimates representative of the true correctness likelihood -- is important for classification models in many applications. We discover that modern neural networks, unlike those from a decade ago, are poorly calibrated. Through extensive experiments, we observe that depth, width, weight decay, and Batch Normalization are important factors influencing calibration. We evaluate the performance of various post-processing calibration methods on state-of-the-art architectures with image and document classification datasets. Our analysis"},"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":"1706.04599","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-06-14T17:33:50Z","cross_cats_sorted":[],"title_canon_sha256":"1362670e665485707138a250830dda24197b5e740f274c065b80524de8c7f243","abstract_canon_sha256":"e1181016070f0348067171a48095ddd802010389154f2fe2713a87406dcc8571"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:38:41.507386Z","signature_b64":"Np8N6zjOkT2oI+8iZsBiPIJEyp+sQJkG/fHRKoOnBMNfbHqR3yzv262PBrFuYhCchVdwcX2yCQDOZauE+UgkBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0be1c865760f6c8121bfa10330d6ce9abad7b1e0bd620e6bd0e612eb82f7c454","last_reissued_at":"2026-05-18T00:38:41.506964Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:38:41.506964Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On Calibration of Modern Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Chuan Guo, Geoff Pleiss, Kilian Q. Weinberger, Yu Sun","submitted_at":"2017-06-14T17:33:50Z","abstract_excerpt":"Confidence calibration -- the problem of predicting probability estimates representative of the true correctness likelihood -- is important for classification models in many applications. We discover that modern neural networks, unlike those from a decade ago, are poorly calibrated. Through extensive experiments, we observe that depth, width, weight decay, and Batch Normalization are important factors influencing calibration. We evaluate the performance of various post-processing calibration methods on state-of-the-art architectures with image and document classification datasets. Our analysis"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1706.04599","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":""},"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":"1706.04599","created_at":"2026-05-18T00:38:41.507026+00:00"},{"alias_kind":"arxiv_version","alias_value":"1706.04599v2","created_at":"2026-05-18T00:38:41.507026+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1706.04599","created_at":"2026-05-18T00:38:41.507026+00:00"},{"alias_kind":"pith_short_12","alias_value":"BPQ4QZLWB5WI","created_at":"2026-05-18T12:31:08.081275+00:00"},{"alias_kind":"pith_short_16","alias_value":"BPQ4QZLWB5WICIN7","created_at":"2026-05-18T12:31:08.081275+00:00"},{"alias_kind":"pith_short_8","alias_value":"BPQ4QZLW","created_at":"2026-05-18T12:31:08.081275+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":25,"internal_anchor_count":13,"sample":[{"citing_arxiv_id":"1906.08312","citing_title":"Calibrated Model-Based Deep Reinforcement Learning","ref_index":19,"is_internal_anchor":true},{"citing_arxiv_id":"2605.22891","citing_title":"Pointwise Metrics Mislead: An Evaluation Protocol for Multimodal Inverse Problems","ref_index":19,"is_internal_anchor":true},{"citing_arxiv_id":"2605.23136","citing_title":"A Gaia-linked High-purity QSO Candidate Catalog in Selected Fields with Extinction-binned Calibration and Spectrum-informed Training","ref_index":25,"is_internal_anchor":true},{"citing_arxiv_id":"1907.06916","citing_title":"Single-bit-per-weight deep convolutional neural networks without batch-normalization layers for embedded systems","ref_index":36,"is_internal_anchor":true},{"citing_arxiv_id":"2605.21566","citing_title":"Calibration, Uncertainty Communication, and Deployment Readiness in CKD Risk Prediction: A Framework Evaluation Study","ref_index":21,"is_internal_anchor":true},{"citing_arxiv_id":"2601.00990","citing_title":"Uncertainty-Calibrated Explainable Artificial Intelligence for Fetal Ultrasound Plane Classification: A Systematic Review","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2605.10843","citing_title":"Training-Free Cultural Alignment of Large Language Models via Persona Disagreement","ref_index":11,"is_internal_anchor":true},{"citing_arxiv_id":"2605.16604","citing_title":"R2V Agent: Teaching SLMs When to Ask for Help","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2605.19369","citing_title":"When to Answer and When to Defer: A Decision Framework for Reliable Code Predictions","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"2506.17788","citing_title":"Bayesian Social Deduction with Graph-Informed Language Models","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2509.20786","citing_title":"LiLAW: Lightweight Learnable Adaptive Weighting to Learn Sample Difficulty & Improve Noisy Training","ref_index":7,"is_internal_anchor":true},{"citing_arxiv_id":"2409.12917","citing_title":"Training Language Models to Self-Correct via Reinforcement Learning","ref_index":198,"is_internal_anchor":true},{"citing_arxiv_id":"2602.01548","citing_title":"Photometric Redshift PDFs via Neural Network Classification for DESI Legacy Imaging Surveys and Pan-STARRS","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2605.03515","citing_title":"Scale-Dependent Input Representation and Confidence Estimation for LLMs in Materials Property Prediction","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10378","citing_title":"Uncertainty in Physics and AI: Taxonomy, Quantification, and Validation","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00195","citing_title":"Diversity in Large Language Models under Supervised Fine-Tuning","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03838","citing_title":"TRACE: A Metrologically-Grounded Engineering Framework for Trustworthy Agentic AI Systems in Operationally Critical Domains","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05715","citing_title":"Decodable but Not Corrected by Fixed Residual-Stream Linear Steering: Evidence from Medical LLM Failure Regimes","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10585","citing_title":"Calibration Collapse Under Sycophancy Fine-Tuning: How Reward Hacking Breaks Uncertainty Quantification in LLMs","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08868","citing_title":"MedFormer-UR: Uncertainty-Routed Transformer for Medical Image Classification","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2604.09791","citing_title":"Pioneer Agent: Continual Improvement of Small Language Models in Production","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06032","citing_title":"Ensemble-Based Dirichlet Modeling for Predictive Uncertainty and Selective Classification","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2207.05221","citing_title":"Language Models (Mostly) Know What They Know","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2604.21193","citing_title":"Trust but Verify: Introducing DAVinCI -- A Framework for Dual Attribution and Verification in Claim Inference for Language Models","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00195","citing_title":"Diversity in Large Language Models under Supervised Fine-Tuning","ref_index":4,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BPQ4QZLWB5WICIN7UEBTBVWOTK","json":"https://pith.science/pith/BPQ4QZLWB5WICIN7UEBTBVWOTK.json","graph_json":"https://pith.science/api/pith-number/BPQ4QZLWB5WICIN7UEBTBVWOTK/graph.json","events_json":"https://pith.science/api/pith-number/BPQ4QZLWB5WICIN7UEBTBVWOTK/events.json","paper":"https://pith.science/paper/BPQ4QZLW"},"agent_actions":{"view_html":"https://pith.science/pith/BPQ4QZLWB5WICIN7UEBTBVWOTK","download_json":"https://pith.science/pith/BPQ4QZLWB5WICIN7UEBTBVWOTK.json","view_paper":"https://pith.science/paper/BPQ4QZLW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1706.04599&json=true","fetch_graph":"https://pith.science/api/pith-number/BPQ4QZLWB5WICIN7UEBTBVWOTK/graph.json","fetch_events":"https://pith.science/api/pith-number/BPQ4QZLWB5WICIN7UEBTBVWOTK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BPQ4QZLWB5WICIN7UEBTBVWOTK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BPQ4QZLWB5WICIN7UEBTBVWOTK/action/storage_attestation","attest_author":"https://pith.science/pith/BPQ4QZLWB5WICIN7UEBTBVWOTK/action/author_attestation","sign_citation":"https://pith.science/pith/BPQ4QZLWB5WICIN7UEBTBVWOTK/action/citation_signature","submit_replication":"https://pith.science/pith/BPQ4QZLWB5WICIN7UEBTBVWOTK/action/replication_record"}},"created_at":"2026-05-18T00:38:41.507026+00:00","updated_at":"2026-05-18T00:38:41.507026+00:00"}