1406 articles tagged with Safety/Alignment
We study word-level semantic alignment across four historical stages of Ancient Egyptian. These stages differ in script and orthography, and parallel data are scarce. We jointly tr...
Existing real-world super-resolution (RSR) methods based on generative priors have achieved remarkable progress in producing high-quality and globally consistent reconstructions. H...
Understanding animal species from multimodal data poses an emerging challenge at the intersection of computer vision and ecology. While recent biological models, such as BioCLIP, h...
A U.S. judge said on Tuesday that the Pentagon's blacklisting of Anthropic looked like an effort to punish the artificial intelligence lab for going public with its concerns about ...
Video-based world models offer a powerful paradigm for embodied simulation and planning, yet state-of-the-art models often generate physically implausible manipulations - such as o...
While Vision-Language Models (VLMs) have achieved remarkable performance, their Euclidean embeddings remain limited in capturing hierarchical relationships such as part-to-whole or...
Improving embodied reasoning in multimodal-large-language models (MLLMs) is essential for building vision-language-action models (VLAs) on top of them to readily translate multimod...
High-performance C++ execution engine for LLM red-teaming and prompt engineering. Deploy dynamic jailbreak payloads, bypass alignment guardrails, and utilize free autonomous uncens...
Culture shapes reasoning, values, prioritization, and strategic decision-making, yet large language models (LLMs) often exhibit cultural biases that misalign with target population...
Our method reconstructs 3D worlds from video diffusion models using non-rigid alignment to resolve inherent 3D inconsistencies in the generated sequences.
Pixel-space diffusion has recently re-emerged as a strong alternative to latent diffusion, enabling high-quality generation without pretrained autoencoders. However, standard pixel...
The point of the blackmail exercise was to have something to describe to policymakers—results that are visceral enough to land with people, and make misalignment risk actually sali...
Representation Alignment (REPA) has emerged as a simple way to accelerate Diffusion Transformers training in latent space. At the same time, pixel-space diffusion transformers such...
We propose Reinforcement Learning from Community Feedback (RLCF), a training paradigm that uses large-scale community signals as supervision, and formulate scientific taste learnin...