SLICE: Sequential Likelihood-Free Inference via Conformal E-Values introduces a method for constructing exact anytime-valid confidence sequences for parameters of scientific models that can be simulated but have intractable likelihoods. The approach uses fresh simulations at each observation to conformalize a positive predictable score into a conditional e-value, whose product forms a test martingale that is inverted over the parameter space. The method is shown to be valid for any score, with the oracle likelihood-ratio score maximizing expected log growth, and is demonstrated on synthetic benchmarks, an audit of model routing in an LLM API, and microlensing data, where the resulting confidence sequences remain valid under continuous monitoring while neural scores remain competitive with exact likelihood ratios and application-specific scores.
arXiv:2505. 04608v5 Announce Type: replace-cross Abstract: Responsibly deploying artificial intelligence (AI) / machine learning (ML) systems in high-stakes settings arguably requires not only proof of system reliability, but also continual, post-deployment monitoring to quickly detect and address any unsafe behavior.
The paper introduces Pro-Active Inference (PAI), a method for efficiently estimating a population mean when only a subset of outcomes can be observed. PAI proactively selects which observation to label next from a covariate pool, updating machine‑learning predictions online, and provides finite‑sample unbiasedness, asymptotic validity, and an improved variance estimator. Experiments on eight datasets show that PAI outperforms Sequential Active Inference in most settings, while diagnostics reveal when additional adaptivity is beneficial or detrimental.
arXiv:2408. 08998v4 Announce Type: replace-cross Abstract: Recent advances in machine learning have significantly improved prediction accuracy in various applications.
Predictable Scaling (PS) is a framework that uses small‑scale training proxies to forecast full‑scale model performance, reducing the need for expensive GPU‑hourly runs. By applying composable scaling functions and curve fitting with uncertainty estimates, PS can rank candidate changes accurately across different hypothesis types. Experiments on two ranking models show PS achieves 79–88% compute savings while maintaining correct ranking of actionable hypotheses.
Neural Composite Likelihood Estimation (NCLE) extends simulation‑based inference to high‑dimensional time series by partitioning long sequences into equal‑sized batches. For each batch, a neural network estimates the likelihood via conditional density estimation, and the product of these batch likelihoods forms an approximate composite likelihood. Frequentist inference is then performed by maximizing this composite likelihood to obtain a point estimate and by estimating the Godambe information matrix to derive confidence intervals.
arXiv:2606. 27711v1 Announce Type: cross Abstract: We introduce a neural network-based framework for learning time series estimators through a process we term decision-theoretic pretraining.
arXiv:2606. 07616v1 Announce Type: cross Abstract: Scaling laws provide a fundamental framework for understanding the performance of Language Models (LMs), yet deriving them requires prohibitively expensive evaluations across thousands of checkpoints or millions of inference samples.
arXiv:2605. 31278v2 Announce Type: replace-cross Abstract: Reliable evaluation of agentic systems requires unbiased estimates with valid uncertainty, but standard practice navigates between costly human annotation and biased LLM-as-judge proxies.
The paper introduces an adaptive, tuning‑free method for weighting the score‑matching term in the loss function of neural ratio estimators, eliminating the need for manual hyper‑parameter tuning. By adjusting the score loss weights during training based on loss gradients, the approach adds minimal overhead to standard classifier training. Experiments on network dynamics and spatial process models show that this technique improves surrogate quality and can match the inference performance of a 10× larger training set while keeping training time under 1.1× higher.
Parallel test-time scaling samples many reasoning traces and majority-votes their answers, improving LLM accuracy but requiring traces to run to completion, incurring substantial computational overhead. We observe that probing partial traces at intermediate checkpoints can extract current answers without disrupting generation, revealing an evolving aggregate vote.
arXiv:2606. 31915v1 Announce Type: cross Abstract: While conformal prediction provides a general framework for uncertainty quantification in predictive inference, its application is often limited by computational cost.
arXiv:2606. 12935v1 Announce Type: new Abstract: Parallel test-time scaling samples many reasoning traces and majority-votes their answers, improving LLM accuracy but requiring traces to run to completion, incurring substantial computational overhead.