Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv:2603. 19703v2 Announce Type: replace-cross Abstract: Estimating covariance matrices is fundamental to a wide range of statistical applications.
arXiv:2608. 12395v1 Announce Type: new Abstract: We describe Research Assistant, an internal LLM-based system developed at AstraZeneca to help scientists and clinicians explore biomedical questions across a broad range of data sources.
arXiv:2608. 12791v1 Announce Type: cross Abstract: What a finite learning device has recorded and what will hold value for it on future tasks are not the same quantity.
arXiv:2608. 12675v1 Announce Type: new Abstract: Retrieval-Augmented Generation (RAG) is widely used to improve the performance of Large Language Models (LLMs) in answering user queries.
arXiv:2608. 12329v1 Announce Type: cross Abstract: Progress on AI for psychosis-risk assessment is limited by a data-access bottleneck.
arXiv:2608. 12355v1 Announce Type: cross Abstract: Recent progress in AI coding agent research has led to rapid improvements in agents' ability to autonomously perform complex software engineering tasks, from editing large codebases to executing long-horizon development workflows.
arXiv:2608. 13221v1 Announce Type: new Abstract: The evaluation of LLM reasoning is moving from final-answer accuracy to process-level assessment, yet existing methods still fail to capture how models plan reasoning paths and allocate reasoning resources--that is, how they organize search.
arXiv:2608. 13345v1 Announce Type: new Abstract: Artificial Intelligence (AI) safety systems combine character shaping (e.
arXiv:2604. 17299v3 Announce Type: replace-cross Abstract: Aligning large language models with human preferences must balance two competing goals: responding helpfully to legitimate requests and reliably refusing harmful ones.
arXiv:2608. 13190v1 Announce Type: new Abstract: Group-robust learning is crucial for maintaining accuracy on rare subpopulations when training-group labels are unavailable.
arXiv:2608. 12373v1 Announce Type: new Abstract: Large language models are increasingly used in strategic and advisory contexts, yet their safety alignment is typically evaluated in English only.
arXiv:2608. 12779v1 Announce Type: cross Abstract: Understanding the temporal progression of symptoms in clinical narratives is critical for disease monitoring, safety surveillance, and causality assessment.
arXiv:2608. 13341v1 Announce Type: cross Abstract: Infrared (IR) spectroscopy is widely used for chemical sensing, but extracting reliable chemical information from spectra remains challenging.
arXiv:2608. 12585v1 Announce Type: new Abstract: Improving reasoning LLMs requires the ability to judge the quality of long reasoning traces for effective reasoning data curation, strong training signals during reinforcement learning, and an in-depth understanding of reasoning behaviors during model performance evaluation.
arXiv:2608. 12372v1 Announce Type: new Abstract: AI systems are increasingly employed as decision aids, decision delegates, or autonomous decision-makers.
arXiv:2608. 13209v1 Announce Type: cross Abstract: Many operational decisions are sequences of interventions under a cumulative resource limit, such as a maintenance schedule within a crew-hour budget.
arXiv:2608. 12876v1 Announce Type: cross Abstract: Detecting AI-generated images is only half the task: a deployed detector must also justify its verdict, yet existing detectors inherit three failure modes from their training data: real and fake images collected from different sources invite provenance shortcuts, supervised explanation corpora teach templated rationales, and a static forgery corpus leaves the decision boundary standing still while generators keep moving.
arXiv:2608. 13118v1 Announce Type: new Abstract: Verification of neural networks against relational specifications, such as global robustness, is crucial for safety-critical applications of cyber-physical systems (CPS), given their increasing adoption of AI components.
arXiv:2508. 14390v2 Announce Type: replace-cross Abstract: Large language models (LLMs) often express verbal confidence that is poorly aligned with actual correctness, limiting their reliability in safety-critical applications.
arXiv:2602. 18733v2 Announce Type: replace Abstract: Training data leakage from Large Language Models (LLMs) raises serious concerns related to privacy, security, and copyright compliance.