English
Related papers

Related papers: LLMs are Bayesian, In Expectation, Not in Realizat…

200 papers

Learning a Bayesian network (BN) from data can be useful for decision-making or discovering causal relationships. However, traditional methods often fail in modern applications, which exhibit a larger number of observed variables than data…

Computation · Statistics 2018-06-26 Raj Agrawal , Tamara Broderick , Caroline Uhler

Before adopting a new large language model (LLM) architecture, it is critical to understand vulnerabilities accurately. Existing evaluations can be difficult to trust, often drawing conclusions from LLMs that are not meaningfully…

Cryptography and Security · Computer Science 2025-10-08 Mary Llewellyn , Annie Gray , Josh Collyer , Michael Harries

Although deep neural networks have demonstrated significant success due to their powerful expressiveness, most models struggle to meet practical requirements for uncertainty estimation. Concurrently, the entangled nature of deep neural…

Machine Learning · Computer Science 2025-05-29 Xinyue Hu , Zhibin Duan , Bo Chen , Mingyuan Zhou

The nondeterminism of Deep Learning (DL) training algorithms and its influence on the explainability of neural network (NN) models are investigated in this work with the help of image classification examples. To discuss the issue, two…

Machine Learning · Computer Science 2022-03-03 A. -M. Leventi-Peetz , T. Östreich

Detecting sandbagging--the deliberate underperformance on capability evaluations--is an open problem in AI safety. We tested whether symptom validity testing (SVT) logic from clinical malingering detection could identify sandbagging through…

Computation and Language · Computer Science 2026-04-29 Jon-Paul Cacioli

Monitoring complex industrial assets relies on engineer-authored symbolic rules that trigger based on sensor conditions and prompt technicians to perform corrective actions. The bottleneck is not detection but response: translating rules…

We provide a reason for Bayesian updating, in the Bernoulli case, even when it is assumed that observations are independent and identically distributed with a fixed but unknown parameter $\theta_0$. The motivation relies on the use of loss…

Statistics Theory · Mathematics 2010-06-08 Pier Giovanni Bissiri , Stephen G. Walker

We study the extent to which standard machine learning algorithms rely on exchangeability and independence of data by introducing a monotone adversarial corruption model. In this model, an adversary, upon looking at a "clean" i.i.d.…

Machine Learning · Computer Science 2026-01-06 Kasper Green Larsen , Chirag Pabbaraju , Abhishek Shetty

Identifying latent representations or causal structures is important for good generalization and downstream task performance. However, both fields have been developed rather independently. We observe that several methods in both…

Machine Learning · Statistics 2025-02-11 Patrik Reizinger , Siyuan Guo , Ferenc Huszár , Bernhard Schölkopf , Wieland Brendel

While pre-trained language models (LMs) have brought great improvements in many NLP tasks, there is increasing attention to explore capabilities of LMs and interpret their predictions. However, existing works usually focus only on a certain…

Computation and Language · Computer Science 2022-07-29 Yaozong Shen , Lijie Wang , Ying Chen , Xinyan Xiao , Jing Liu , Hua Wu

Large language models adapt to new tasks through in-context learning (ICL) without parameter updates. Current theoretical explanations for this capability assume test tasks are drawn from a distribution similar to that seen during…

Machine Learning · Computer Science 2026-02-23 Di Zhang

Video anomaly detection is a challenging task because most anomalies are scarce and non-deterministic. Many approaches investigate the reconstruction difference between normal and abnormal patterns, but neglect that anomalies do not…

Computer Vision and Pattern Recognition · Computer Science 2022-06-28 Guodong Shen , Yuqi Ouyang , Victor Sanchez

We argue for supplementing the process of training a prediction algorithm by setting up a scheme for detecting the moment when the distribution of the data changes and the algorithm needs to be retrained. Our proposed schemes are based on…

Machine Learning · Computer Science 2021-02-23 Vladimir Vovk , Ivan Petej , Ilia Nouretdinov , Ernst Ahlberg , Lars Carlsson , Alex Gammerman

Standard LLM evaluations only test capabilities or dispositions that evaluators designed them for, missing unexpected differences such as behavioral shifts between model revisions or emergent misaligned tendencies. Model diffing addresses…

Machine Learning · Computer Science 2026-02-12 Elias Kempf , Simon Schrodi , Bartosz Cywiński , Thomas Brox , Neel Nanda , Arthur Conmy

Multimodal large language models (MLLMs) have shown promising capabilities but struggle under distribution shifts, where evaluation data differ from instruction tuning distributions. Although previous works have provided empirical…

Artificial Intelligence · Computer Science 2025-05-27 Changdae Oh , Zhen Fang , Shawn Im , Xuefeng Du , Yixuan Li

In recent years, pre-trained large language models (LLMs) have demonstrated remarkable efficiency in achieving an inference-time few-shot learning capability known as in-context learning. However, existing literature has highlighted the…

Computation and Language · Computer Science 2024-02-14 Xinyi Wang , Wanrong Zhu , Michael Saxon , Mark Steyvers , William Yang Wang

This paper addresses learning stochastic rules especially on an inter-attribute relation based on a Minimum Description Length (MDL) principle with a finite number of examples, assuming an application to the design of intelligent relational…

Artificial Intelligence · Computer Science 2013-03-08 Joe Suzuki

Successful machine learning methods require a trade-off between memorization and generalization. Too much memorization and the model cannot generalize to unobserved examples. Too much over-generalization and we risk under-fitting the data.…

Artificial Intelligence · Computer Science 2023-03-09 Chase Yakaboski , Eugene Santos

Deep neural networks can obtain impressive performance on various tasks under the assumption that their training domain is identical to their target domain. Performance can drop dramatically when this assumption does not hold. One…

Machine Learning · Computer Science 2024-10-10 Gaël Gendron , Michael Witbrock , Gillian Dobbie

Max-Linear Bayesian Networks (MLBNs) provide a powerful framework for causal inference in extreme-value settings; we consider MLBNs with noise parameters with a given topology in terms of the max-plus algebra by taking its logarithm. Then,…

Machine Learning · Statistics 2025-05-02 Mark Adams , Kamillo Ferry , Ruriko Yoshida