Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv:2609.36756v1 Announce Type: cross Abstract: One-dimensional (1D) variable-length visual tokenizers enable adaptive compression by varying the number of tokens, allowing downstream autoregressiv...
arXiv:2609.36915v1 Announce Type: cross Abstract: Aerial manipulators extend robotic manipulation into 3D workspaces that are difficult for ground-based robots to access, creating new opportunities f...
arXiv:2609.36942v1 Announce Type: cross Abstract: Learning neural-network models of dynamical systems with safety guarantees is a fundamental requirement for their deployment in safety-critical setti...
arXiv:2609.36952v1 Announce Type: cross Abstract: Large language models (LLMs) excel at token-level generation but may learn undesirable abstract semantics and lack comprehensive perception. LLM-JEPA...
arXiv:2609.36995v1 Announce Type: cross Abstract: Few-step streaming audio--video generation requires both causal modeling and step distillation, yet standard training recipes face two context-relate...
arXiv:2609.37030v1 Announce Type: cross Abstract: Despite rapid progress in video generation models, they still exhibit obvious motion deficiencies, often manifested as incorrect object motion. Howev...
arXiv:2609.37061v1 Announce Type: cross Abstract: We view fairness as a property of distributional stability. Rather than assessing a predictor under a fixed data distribution, we study how its predi...
arXiv:2609.37119v1 Announce Type: cross Abstract: Recent approaches to reinforcement learning (RL) post-training for large language models increasingly remove the critic to reduce training instabilit...
arXiv:2609.37243v1 Announce Type: cross Abstract: Cross-modal knowledge distillation transfers knowledge from a teacher modality to a student modality. Existing feature-level alignment methods typica...
arXiv:2609.37344v1 Announce Type: cross Abstract: Data reconstruction attacks have empirically been successful in recovering training samples from learned models, raising privacy concerns and motivat...
arXiv:2609.37424v1 Announce Type: cross Abstract: Optimal Transport (OT) provides a principled framework for learning transformations between probability distributions from unpaired samples. In many...
arXiv:2609.37501v1 Announce Type: cross Abstract: We propose RegLLM, a diagnostic harness for bounded autonomy in regulated agentic workflows. It instruments six trustworthiness signals: citation val...
arXiv:2609.37669v1 Announce Type: cross Abstract: Retrieval-Augmented Generation (RAG) is increasingly used to enhance Large Language Model (LLM)-based software vulnerability detection by grounding p...
arXiv:2609.37871v1 Announce Type: cross Abstract: Average performance on routine driving benchmarks does not establish planner reliability under rare, safety-critical hazards. We proposed ExceptionDr...
arXiv:2609.37914v1 Announce Type: cross Abstract: Fine-tuning large language models on narrow, misaligned tasks can undo their post-training alignment and induce novel misaligned behaviors -- a pheno...
arXiv:2609.37938v1 Announce Type: cross Abstract: Embodied systems must make knowledge acquired during one encounter usable in another despite changes in viewpoint, motion, and illumination. Yet aggr...
arXiv:2609.37974v1 Announce Type: cross Abstract: Masked diffusion models (MDMs) generate text by unmasking several tokens per step, but they are trained and sampled under different conditions. The m...
arXiv:2609.38010v1 Announce Type: cross Abstract: Modern computer vision models achieve high accuracy when trained on large-scale annotated datasets. In critical domains such as construction safety m...
arXiv:2609.38107v1 Announce Type: cross Abstract: Chain-of-thought traces are widely read as records of how models reach their answers, informing debugging, agent auditing, and claims about reasoning...
arXiv:2609.38109v1 Announce Type: cross Abstract: The attention operation is naively position invariant. However, positional information is fundamental to natural language, and therefore a variety of...