SLICE: Sequential Likelihood-Free Inference via Conformal E-Values
stat.ME, stat.ML
Submitted: 2026-10-09
Updated: 2026-10-13
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- Are You Getting What You Pay For? Auditing Model Substitution in LLM APIs
- Approximating Likelihood Ratios with Calibrated Discriminative Classifiers
- Besag-Clifford e-values for unnormalized testing
- Anytime-valid simulation-based hypothesis testing
- Computing Linear Restrictions of Neural Networks
- Representation Learning with Contrastive Predictive Coding
- M-estimation with e-statistics
- Qwen3 Technical Report
- Which Model Is Actually Serving You? IRIS: Budgeted Black-Box Auditing of Model Substitution and Routing Dilution in LLM Gateways
Related papers
- Doubly robust inference via calibration
- Bayesian Empirical Bayes: Simultaneous Inference from Probabilistic Symmetries
- Optimal Watermark Localization in Mixed-Source Large Language Model Texts
- Flexible Nonparametric Inference for Causal Effects under the Front-Door Model
- A Greedy PDE Router for Blending Neural Operators and Classical Methods
- Policy Learning with Weak Signals