Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
With the rapid development of Large language model (LLM), agent systems enhanced by LLMs show huge potential in being able to deal with complex tasks, especially involving multi-step thinking or inter...
We give deterministic and uncoupled learning dynamics for finite multiplayer general-sum games under full-information feedback that achieve constant individual swap regret, independent of the horizon...
Category-level object pose estimation seeks to recover a similarity transform $(R,t,s)$ for unseen instances without instance-specific CAD models. Most competitive methods are correspondence-based: pr...
Large-scale vision-language models (VLM) such as CLIP enable strong open-vocabulary reasoning, yet deploying these capabilities on resource-constrained edge devices remains challenging. EdgeVL address...
Leveraging Large Vision-Language Models like CLIP has recently set new benchmarks for No-Reference Image Quality Assessment (NR-IQA). However, the contrastive pretraining of CLIP inherently prioritize...
Neural PDE solvers provide efficient surrogates for time-dependent physical systems, but autoregressive prediction over long horizons remains challenging because local errors can induce distribution s...
The article "Math for AI safety: an invitation for mathematicians" calls for new mathematical tools to ensure AI remains understandable, controllable, and cooperative. It outlines specific mathematical fields—logic, game theory, probability, algebra, representation theory, analysis, and geometry—each paired with an open problem tailored for mathematicians without AI safety background. The piece invites researchers to contribute to designing AI that is legible, steerable, and aligned with human values.
arXiv:2609.13984v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models combine a pretrained vision encoder, a language backbone, and an action head, but their relative contribution has...
arXiv:2602.21061v2 Announce Type: replace Abstract: Many current paths to more advanced AI depend on the assumption that large language models (LLMs) can generalize learned relationships to solve com...
arXiv:2609.13734v1 Announce Type: cross Abstract: As large language models (LLMs) are increasingly deployed in real-world high-stakes applications, effective governance has become essential. Existing...
arXiv:2609.14595v1 Announce Type: cross Abstract: Multichannel time-series classification commonly assumes synchronized sensor streams, although latency, clock drift, and preprocessing can introduce...
arXiv:2609.15533v1 Announce Type: cross Abstract: Mechanistic interpretability has become the dominant lens for understanding frontier language models, as their inner workings are complex and inheren...
arXiv:2609.15488v1 Announce Type: new Abstract: Machine learning models process large amounts of data, and Principal Component Analysis (PCA) is a widely used technique to reduce the dimensionality o...
arXiv:2609.15915v1 Announce Type: new Abstract: Meta-reinforcement learning (meta-RL) enables agents to adapt to unseen tasks with limited experience. Despite its promise, the application of meta-RL...
arXiv:2603.18908v5 Announce Type: replace Abstract: Independently trained language models often learn compatible late-stage representations, despite differences in training objectives, architectures,...
arXiv:2603.19896v2 Announce Type: replace Abstract: Tool-using large language model (LLM) agents often face a fundamental tension between answer quality and execution cost. Fixed workflows are stable...
arXiv:2609.14796v1 Announce Type: new Abstract: The threat that AI persuasion poses to human control has been acknowledged in the literature, but not yet systematically studied. Now that persuasion a...
arXiv:2609.14449v1 Announce Type: new Abstract: Recent work has shown that a teacher model can transfer a bias to a student through a dataset from which every explicit reference to that bias has been...
arXiv:2603.18740v3 Announce Type: replace-cross Abstract: Automated Code Review (ACR) systems integrating Large Language Models (LLMs) are increasingly adopted in software development workflows, rang...
ViCo-SAM3 introduces a Vision-Conditioned alignment framework for open-vocabulary camouflaged object segmentation. The approach adds a vision-conditioned (ViCo) module that dynamically adjusts text embeddings based on global visual context, and a vision-conditioned cross-modal binding (ViCoBind) module to improve interaction between visual and textual representations. These innovations close the semantic gap between text and pixel-level cues, enabling state‑of‑the‑art performance on the OVCamo benchmark without heavy parameter overhead.