arXiv Machine Learning July 22, 2026 By Garrett Baker, Vinayak Pathak, Daniel Murfet, Susan Wei

Elicitation without Backpropagation: Steering Model Behavior by Optimizing the Latent Posterior

Read the original on arXiv Machine Learning →

arXiv:2607. 18804v1 Announce Type: new Abstract: In the \emph{latent posterior model} of transformer behavior, the next-token distribution arises from a posterior over latent predictive models conditioned on the context, mixed to generate continuations.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

llms

arXiv Machine Learning

Jul 21

What does a Bayes-filtered transformer believe? A predictive Monte Carlo approach

arXiv:2607. 17060v1 Announce Type: new Abstract: A Bayes-filtered transformer (BFT) is a transformer trained on sequences that are generated in two steps: first a latent task is drawn from a prior, then observations are drawn conditional on that task.

By Afiq Abdillah Effiezal Aswadi, Haotong Ma, Susan Wei

llms diffusion safety

arXiv Machine Learning

Jun 24

LLMs are Bayesian, In Expectation, Not in Realization

arXiv:2507. 11768v3 Announce Type: replace-cross Abstract: Bayesian accounts of in-context learning face a direct objection: exact posterior predictives for exchangeable data are invariant to task-preserving order, yet transformers change next-token probabilities when the same examples are serialized differently.

By Leon Chlon, Fatima Sheaib, Zein Khamis, Maggie Chlon, Mahdi El Zein, MarcAntonio M. Awada

llms benchmarks

arXiv AI

Jul 1

Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation

arXiv:2606. 31184v1 Announce Type: cross Abstract: Adaptive experiments for average treatment effects (ATE) require randomized allocations balancing valid inference with statistical efficiency.

By Jiachun Li, David Simchi-Levi

llms

arXiv Machine Learning

Jun 18

Decomposing Prediction Mechanisms for In-Context Recall

arXiv:2507. 01414v2 Announce Type: replace Abstract: We introduce a new family of toy problems that combine features of linear-regression-style continuous in-context learning (ICL) with discrete associative recall.

By Sultan Daniels, Dylan Davis, Dhruv Gautam, Wentinn Liao, Gireeja Ranade, Anant Sahai

llms efficiency

arXiv Machine Learning

Jul 21

Uncertainty Decomposition for Bayes-Filtered Transformers via Bayesian Predictive Inference

arXiv:2602. 04596v2 Announce Type: replace-cross Abstract: Bayes-filtered transformers are transformers meta-learned on sequences from a prior predictive distribution to approximate the corresponding posterior predictive distribution.

By Sandra Fortini, Kenyon Ng, Sonia Petrone, Judith Rousseau, Susan Wei

llms benchmarks

arXiv Machine Learning

Jul 7

Gradient Descent as Implicit EM in Distance-Based Neural Models

arXiv:2512. 24780v2 Announce Type: replace Abstract: Neural networks trained with standard objectives exhibit behaviors characteristic of probabilistic inference: soft clustering, prototype specialization, and Bayesian uncertainty tracking.

By Alan Oursland

llms

One email a morning, machine-written

Related stories

What does a Bayes-filtered transformer believe? A predictive Monte Carlo approach

LLMs are Bayesian, In Expectation, Not in Realization

Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation

Decomposing Prediction Mechanisms for In-Context Recall

Uncertainty Decomposition for Bayes-Filtered Transformers via Bayesian Predictive Inference

Gradient Descent as Implicit EM in Distance-Based Neural Models