English
Related papers

Related papers: Predicting Where Steering Vectors Succeed

200 papers

While machine learning systems show high success rate in many complex tasks, research shows they can also fail in very unexpected situations. Rise of machine learning products in safety-critical industries cause an increase in attention in…

Machine Learning · Computer Science 2019-05-21 Sina Mohseni , Akshay Jagadeesh , Zhangyang Wang

Understanding and predicting highway lane-change maneuvers is essential for driving modeling and its automation. The development of data-based lane-changing decision-making algorithms is nowadays in full expansion. We compare empirically in…

Machine Learning · Computer Science 2023-02-08 Basma Khelfa , Ibrahima Ba , Antoine Tordeux

Modern AI models contain much of human knowledge, yet understanding of their internal representation of this knowledge remains elusive. Characterizing the structure and properties of this representation will lead to improvements in model…

Computation and Language · Computer Science 2025-05-30 Daniel Beaglehole , Adityanarayanan Radhakrishnan , Enric Boix-Adserà , Mikhail Belkin

Averaging neural network parameters is an intuitive method for fusing the knowledge of two independent models. It is most prominently used in federated learning. If models are averaged at the end of training, this can only lead to a good…

Machine Learning · Computer Science 2024-03-20 Linara Adilova , Maksym Andriushchenko , Michael Kamp , Asja Fischer , Martin Jaggi

Calibration models have been developed for determination of trace elements, silver for instance, in soil using laser-induced breakdown spectroscopy (LIBS). The major concern is the matrix effect. Although it affects the accuracy of LIBS…

Deploying machine learning models in safety-critical domains poses a key challenge: ensuring reliable model performance on downstream user data without access to ground truth labels for direct validation. We propose the suitability filter,…

Machine Learning · Computer Science 2025-05-29 Angéline Pouget , Mohammad Yaghini , Stephan Rabanser , Nicolas Papernot

Instruction-based image editing exhibits heterogeneous difficulty not only across cases but also across regions of an image, motivating refinement approaches that allocate correction to where the model struggles. Existing refinement signals…

Computer Vision and Pattern Recognition · Computer Science 2026-05-14 Weitai Kang , Xiaohang Zhan , Yizhou Wang , Mang Tik Chiu , Jason Kuen , Kangning Liu , Yan Yan

In traditional logistic regression models, the link function is often assumed to be linear and continuous in predictors. Here, we consider a threshold model that all continuous features are discretized into ordinal levels, which further…

Methodology · Statistics 2022-02-18 Yinan Lin , Wen Zhou , Zhi Geng , Gexin Xiao , Jianxin Yin

Neural network classifiers trained on datasets with uneven group representation often inherit class biases and learn spurious correlations. These models may perform well on average but consistently fail on atypical groups. For example, in…

Machine Learning · Computer Science 2025-06-24 Aviral Gupta , Armaan Sethi , Ameesh Sethi

Lane detection is a critical component of Advanced Driver-Assistance Systems (ADAS) and Automated Driving System (ADS), providing essential spatial information for lateral control. However, domain shifts often undermine model reliability…

Computer Vision and Pattern Recognition · Computer Science 2025-07-18 Yin Wu , Daniel Slieter , Ahmed Abouelazm , Christian Hubschneider , J. Marius Zöllner

We present the Latent Adaptive Planner (LAP), a trajectory-level latent-variable policy for dynamic nonprehensile manipulation (e.g., box catching) that formulates planning as inference in a low-dimensional latent space and is learned…

Robotics · Computer Science 2025-11-25 Donghun Noh , Deqian Kong , Minglu Zhao , Andrew Lizarraga , Jianwen Xie , Ying Nian Wu , Dennis Hong

We study the estimation of the linear discriminant with projection pursuit, a method that is blind in the sense that it does not use the class labels in the estimation. Our viewpoint is asymptotic and, as our main contribution, we derive…

Statistics Theory · Mathematics 2023-08-15 Una Radojicic , Klaus Nordhausen , Joni Virta

Robust autonomy stacks require tight integration of perception, motion planning, and control layers, but these layers often inadequately incorporate inherent perception and prediction uncertainties, either ignoring them altogether or making…

Systems and Control · Electrical Eng. & Systems 2022-01-06 Venkatraman Renganathan , Sleiman Safaoui , Aadi Kothari , Benjamin Gravell , Iman Shames , Tyler Summers

Linear probes can detect when language models produce outputs they "know" are wrong, a capability relevant to both deception and reward hacking. However, single-layer probes are fragile: the best layer varies across models and tasks, and…

Machine Learning · Computer Science 2026-04-16 Erik Nordby , Tasha Pais , Aviel Parrack

Deeply-learned planning methods are often based on learning representations that are optimized for unrelated tasks. For example, they might be trained on reconstructing the environment. These representations are then combined with predictor…

Machine Learning · Computer Science 2021-03-18 Hlynur Davíð Hlynsson , Merlin Schüler , Robin Schiewer , Tobias Glasmachers , Laurenz Wiskott

The introduction of pre-trained transformer-based contextualized word embeddings has led to considerable improvements in the accuracy of graph-based parsers for frameworks such as Universal Dependencies (UD). However, previous works differ…

Computation and Language · Computer Science 2021-07-30 Stefan Grünewald , Annemarie Friedrich , Jonas Kuhn

Most existing vision-language-action (VLA) models for robotic manipulation lack progress awareness, typically relying on hand-crafted heuristics for task termination. This limitation is particularly severe in long-horizon tasks involving…

Robotics · Computer Science 2026-03-31 Hongyu Yan , Qiwei Li , Jiaolong Yang , Yadong Mu

Model steering, which involves intervening on hidden representations at inference time, has emerged as a lightweight alternative to finetuning for precisely controlling large language models. While steering efficacy has been widely studied,…

Machine Learning · Computer Science 2026-02-09 Navita Goyal , Hal Daumé

Controlling the behavior of large language models (LLMs) at inference time is essential for aligning outputs with human abilities and safety requirements. \emph{Activation steering} provides a lightweight alternative to prompt engineering…

Artificial Intelligence · Computer Science 2026-01-30 Diaoulé Diallo , Katharina Dworatzyk , Sophie Jentzsch , Peer Schütt , Sabine Theis , Tobias Hecking

Set classification aims to classify a set of observations as a whole, as opposed to classifying individual observations separately. To formally understand the unfamiliar concept of binary set classification, we first investigate the optimal…

Machine Learning · Statistics 2020-06-29 Zhao Ren , Sungkyu Jung , Xingye Qiao