The minimax rate of estimating second-order calibration error is Õ(1/√n) with a matching Ω(1/√n) lower bound, enabled by analyticity from the sech kernel and yielding the first finite-sample guarantee for second-order Platt scaling.
Proceedings of the AAAI conference on artificial intelligence , volume=
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 5roles
background 1polarities
background 1representative citing papers
Risk-controlled post-processing yields a threshold-structured policy that follows the baseline except where an oracle fallback sharply reduces conditional violation risk, achieving O(log n/n) expected excess risk in i.i.d. settings and exact risk control under exchangeability.
Geometric features from per-layer MLP update trajectories fed to a sparse linear probe outperform maximum softmax probability for uncertainty quantification under selective abstention, with gains up to 21 AURC points.
Conflicting biomedical evidence triggers order-dependent prediction flips in RAG LLMs, and a new abstention score combining confidence with conflict detection raises selective accuracy by 7-33 points in the hardest conditions.
A reference-free proxy scoring framework combined with GIRB calibration produces better-aligned evaluation metrics for summarization and outperforms baselines across seven datasets.
citing papers explorer
-
The Minimax Rate of Second-Order Calibration
The minimax rate of estimating second-order calibration error is Õ(1/√n) with a matching Ω(1/√n) lower bound, enabled by analyticity from the sech kernel and yielding the first finite-sample guarantee for second-order Platt scaling.
-
Risk-Controlled Post-Processing of Decision Policies
Risk-controlled post-processing yields a threshold-structured policy that follows the baseline except where an oracle fallback sharply reduces conditional violation risk, achieving O(log n/n) expected excess risk in i.i.d. settings and exact risk control under exchangeability.
-
Reading Calibrated Uncertainty from Language Model Trajectories
Geometric features from per-layer MLP update trajectories fed to a sparse linear probe outperform maximum softmax probability for uncertainty quantification under selective abstention, with gains up to 21 AURC points.
-
When Evidence Conflicts: Uncertainty and Order Effects in Retrieval-Augmented Biomedical Question Answering
Conflicting biomedical evidence triggers order-dependent prediction flips in RAG LLMs, and a new abstention score combining confidence with conflict detection raises selective accuracy by 7-33 points in the hardest conditions.
-
Calibrating Model-Based Evaluation Metrics for Summarization
A reference-free proxy scoring framework combined with GIRB calibration produces better-aligned evaluation metrics for summarization and outperforms baselines across seven datasets.