Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv:2608.28629v1 Announce Type: cross Abstract: Existing methods lack a generalized approach to efficiently identify and resolve the diversity of design defects in BIM. Therefore, this study propos...
arXiv:2608.28645v1 Announce Type: cross Abstract: Low-resource languages without an adequate training corpus often use a related, higher-resource language as a scaffold for comprehension. Still, ther...
arXiv:2608.28802v1 Announce Type: cross Abstract: Facial appearance editing powers popular applications like FaceApp and Photoshop. Generative Adversarial Networks (GANs) and 3D Morphable Models (3DM...
arXiv:2608.28859v1 Announce Type: cross Abstract: Reasoning models do not stop when they know the answer. On DeepSeek-R1-Distill-Qwen-7B the chain of thought runs about twice as long as the model's o...
arXiv:2608.29034v1 Announce Type: cross Abstract: A wide range of methods have been proposed for interpreting language models, delivering important insights into their inner workings. However, differ...
arXiv:2608.29111v1 Announce Type: cross Abstract: Applications such as personalized assistance and proprietary document analysis require large language models (LLMs) to generate outputs from private...
arXiv:2608.29136v1 Announce Type: cross Abstract: We ask whether a model protects a user in the same way when that user speaks rather than types. Using a single distress vignette---a physical injury...
arXiv:2608.29179v1 Announce Type: cross Abstract: Generative recommendation encodes items as hierarchical semantic identifiers (SIDs) and retrieves the next item through autoregressive decoding. Stan...
arXiv:2608.29266v1 Announce Type: cross Abstract: Researchers increasingly treat LLM survey responses as a proxy for human cultural values. This includes projecting model outputs onto instruments lik...
arXiv:2608.29378v1 Announce Type: cross Abstract: Arabic large language models must refuse harmful prompts without over-refusing benign or sensitive prompts, yet a single refusal rate hides this trad...
arXiv:2608.29458v1 Announce Type: cross Abstract: Sandbagging, in which a model deliberately underperforms on an evaluation despite retaining the underlying capability, threatens the safety evaluatio...
arXiv:2608.29464v1 Announce Type: cross Abstract: Chain-of-thought (CoT) monitoring assumes that reasoning traces faithfully record the information that shapes a model's answer. Existing faithfulness...
arXiv:2608.29490v1 Announce Type: cross Abstract: Multi-agent systems in the real-world (e.g., drone swarms, autonomous cars, warehouse robots) must satisfy rich, temporal tasks while avoiding collis...
arXiv:2608.29563v1 Announce Type: cross Abstract: School coaches prepare for opponents with game film and intuition. The analytics tools of professional teams stay out of reach. We ask how far public...
arXiv:2608.29582v1 Announce Type: cross Abstract: Current evaluations of large language models (LLMs) primarily focus on factual knowledge retrieval, overlooking the fundamental challenge of navigati...
arXiv:2608.29623v1 Announce Type: cross Abstract: Recent advances in large reasoning models (LRMs) have shown strong performance on complex problems through long chain-of-thought (Long CoT) reasoning...
arXiv:2608.29924v1 Announce Type: cross Abstract: Large Vision-Language Models (LVLMs) are prone to hallucinations: they fluently describe objects, attributes, and scenes that are not in the image. W...
arXiv:2608.30122v1 Announce Type: cross Abstract: Vision-language-action (VLA) driving methods increasingly combine multi-trajectory imitation learning with group-relative policy optimization (GRPO),...
arXiv:2608.30378v1 Announce Type: cross Abstract: Direct vision-language-action policies generate continuous robot actions efficiently, but standard behavior cloning leaves two complementary gaps: th...
arXiv:2608.30428v1 Announce Type: cross Abstract: Strategic deception by LLM and VLM agents has emerged as a central AI alignment and safety concern. Social-deduction games (where each player holds a...