Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv:2306.00636v3 Announce Type: replace-cross Abstract: Many fairness criteria constrain the policy or choice of predictors, which can have unwanted consequences, in particular, when optimizing the...
arXiv:2404.06023v3 Announce Type: replace-cross Abstract: Motivated by Q-learning, we study nonsmooth contractive stochastic approximation (SA) with constant stepsize. We focus on two important class...
arXiv:2601.00900v2 Announce Type: replace-cross Abstract: As a critical application of computational intelligence in remote sensing, deep learning-based synthetic aperture radar (SAR) image target re...
arXiv:2606.28876v4 Announce Type: replace-cross Abstract: Memory-Mediated Learning Architecture (MMLA) separates slow base parameters theta, a bounded numerical policy carrier Phi, and a bounded auth...
arXiv:2609.13534v1 Announce Type: new Abstract: We identify \textbf{Harmfulness Propagation Dynamics (HPD)}: for harmful prompts, the projection of the last-token hidden state onto a learned harm dir...
arXiv:2609.13709v1 Announce Type: new Abstract: Human corrections identify editable spans, but the examples receiving corrections may come from a selective feedback channel. We analyze this interacti...
arXiv:2609.13768v1 Announce Type: new Abstract: Multi-hop question answering often fails when retrieval treats evidence as isolated matches to the original question, since the facts needed to answer...
arXiv:2609.13824v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used to evaluate the responses of other language models. This approach, known as LLM-as-a-Judge, is faste...
arXiv:2609.14226v1 Announce Type: new Abstract: Automated radiology report generation has advanced rapidly in diagnostic accuracy, yet generated reports frequently diverge from the stylistic conventi...
arXiv:2609.14256v1 Announce Type: new Abstract: Topic models are widely used to analyze public health-related social media short texts, yet their evaluation remains dominated by metrics that focus en...
arXiv:2609.14302v1 Announce Type: new Abstract: Can financial vision-language models (VLMs) turn chart evidence into reliable action recommendations? Existing hallucination evaluations are mostly cla...
arXiv:2609.15022v1 Announce Type: new Abstract: Defense is a knowledge-intensive domain that requires precise understanding of specialized terminology, doctrinal concepts, operational procedures, and...
arXiv:2609.15886v1 Announce Type: new Abstract: Large language models (LLMs) often learn both desirable and undesirable properties during post-training. We study whether midtraining, an earlier train...
arXiv:2609.14738v1 Announce Type: new Abstract: Automated reviewing systems are increasingly evaluated based on the quality of the reviews they produce. Yet a review is only useful if acting on it le...
arXiv:2602.00377v3 Announce Type: replace Abstract: Existing knowledge probing methods rely on pre-defined queries, limiting extraction to known concepts. We introduce DecompressionLM, a stateless fr...
arXiv:2603.29396v2 Announce Type: replace Abstract: Standard evaluations of Large language models (LLMs) focus on task performance, offering limited insight into whether correct behavior reflects app...
arXiv:2604.09629v3 Announce Type: replace Abstract: Humor generation poses a significant challenge for Large Language Models (LLMs), because their standard training objective (next-token prediction)...
arXiv:2606.27314v2 Announce Type: replace Abstract: To avoid moderation and surveillance on social media, some users routinely invent indirect linguistic expressions (ILE) that camouflage sensitive m...
arXiv:2606.00684v3 Announce Type: replace-cross Abstract: We address the problem of out-of-distribution (OOD) detection for target observations embedded in a subspace of the high dimensional data spa...
arXiv:2609.13255v1 Announce Type: new Abstract: Facial state analysis plays a crucial role in understanding human expressions, psychological modeling, and human computer interaction. Traditional unim...