Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv:2602. 07697v3 Announce Type: replace-cross Abstract: Predictive coding (PC) is a biologically plausible alternative to standard backpropagation (BP) that minimises an energy function with respect to network activities before updating weights.
arXiv:2301. 07210v5 Announce Type: replace-cross Abstract: Digital twins are simulation-based models designed to predict how a real-world process will evolve in response to interventions.
arXiv:2308. 04553v4 Announce Type: replace-cross Abstract: Visual recognition models are prone to learning spurious correlations induced by a biased training set where certain conditions $B$ (\eg, Indoors) are over-represented in certain classes $Y$ (\eg, Big Dogs).
arXiv:2510. 23389v2 Announce Type: replace-cross Abstract: The behaviour of neural network components must be proven correct before deployment in safety-critical systems.
arXiv:2511. 19735v2 Announce Type: replace-cross Abstract: Randomized controlled trials (RCTs)have been the cornerstone of clinical evidence; however, their cost, duration, and restrictive eligibility criteria limit power and external validity.
arXiv:2604. 07282v2 Announce Type: replace-cross Abstract: Automated face recognition has made rapid strides over the past decade due to the unprecedented rise of deep neural network (DNN) models that can be trained for domain-specific tasks.
arXiv:2602. 24210v3 Announce Type: replace-cross Abstract: Large reasoning models (LRMs) produce reasoning traces (RTs) that often contain sensitive information.
arXiv:2603. 11398v2 Announce Type: replace-cross Abstract: Recently, visual localization has become an important supplement to improve localization reliability, and cross-view approaches can greatly enhance coverage and adaptability.
arXiv:2604. 11704v2 Announce Type: replace-cross Abstract: Deep Neural Networks are highly susceptible to shortcut learning, frequently memorizing low-dimensional spurious correlations instead of underlying causal mechanisms.
arXiv:2604. 03480v2 Announce Type: replace-cross Abstract: Creative thinking is a fundamental aspect of human cognition, and divergent thinking-the capacity to generate novel and varied ideas-is widely regarded as its core generative engine.
arXiv:2604. 08005v2 Announce Type: replace Abstract: Advancements in multimodal foundation models have enabled the development of Computer Use Agents (CUAs) capable of autonomously interacting with GUI environments.
arXiv:2608. 08288v1 Announce Type: new Abstract: Estimating counterfactual outcomes over time from longitudinal observational data is central to clinical decision support.
arXiv:2608. 09164v1 Announce Type: new Abstract: Aligning large language models (LLMs) with human privacy preferences requires capturing individuals' disclosure boundaries beyond general privacy norms.
arXiv:2608. 08422v1 Announce Type: cross Abstract: Ranking data arise in scientific and machine learning applications, including recommendation systems, information retrieval, voting, marketing, and AI preference ranking from human feedback.
Machine learning-based Intrusion Detection Systems (IDS) have demonstrated superior performance in securing Unmanned Aerial Vehicle (UAV) networks. However, the 'black-box' nature of these models, combined with the high dimensionality of multimodal cyber-physical data, poses significant interpretability challenges.
Forecast accuracy does not tell us which past inputs produced a prediction. We separate three questions for time-series models with known delay structure: can the true delay be recovered from the observed data, does the model report it, and does the forecast actually use the same history?
Deep neural networks are increasingly deployed in safety-critical domains as perception modules, where failures are often caused due to rare and under-represented scenarios. This necessitates the need to evaluate the semantic robustness of perception models; conformance of behavior to high-level requirements over real-world perceptual variability.
Lens methods interpret large language models (LLMs) by mapping intermediate activations to the output vocabulary, revealing how next-token predictions develop through the network. Trained lenses remain expensive: affine-translator parameters grow quadratically with model width, while exact, full-vocabulary Kullback--Leibler (KL) training dominates memory.
Detecting phonemes from children's speech has historically been difficult due to the scarcity of training data, and unique characteristics of children's speech. During a phoneme detection competition, we found that training a lightweight model to predict the age of the learner, as well as the phoneme sequence, enabled a 94M-parameter model to outperform WavLM Large models (317M) on the target DrivenData distribution, and fall within approximately 0.
Video anomaly detection (VAD) aims to identify and temporally localize abnormal events in videos. Supervised methods learn anomaly decision boundaries from target-domain annotations but require substantial in-domain data.