As we climbed the winding road to Bédar, we emerged into a charred and desolate landscape. Molten car parts littered our path and out of the window we saw the whole hillside now coated in a dusty black. At least 13 people, including five believed to be Britons, were killed by Thursday's wildfire in Spain's Almeria province, one of the country's deadliest ever. The toll rose on Sunday after a 93-year-old woman, believed to be British, died of her injuries in hospital. The identities of those killed have not yet been officially confirmed.

Large vision-language models (LVLMs) derive their capabilities from extensive training on vast corpora of visual and textual data.

Compilers, while essential, are notoriously complex systems that demand prohibitively expensive human expertise to develop and maintain. The recent advancements in Large Language Models (LLMs) offer a compelling new paradigm: Neural Compilation, which could potentially simplify compiler development for new architectures and facilitate the discovery of innovative optimization techniques. However, several critical obstacles impede its practical adoption. Firstly, a significant lack of dedicated benchmarks and robust evaluation methodologies hinders objective assessment and tracking of progress in the field. Secondly, systematically enhancing the reliability and performance of LLM-generated assembly remains a critical challenge.

Text-to-image (T2I) diffusion models have made remarkable strides in generating and editing high-fidelity images from text. Yet, these models remain fundamentally generic, failing to adapt to the nuanced aesthetic preferences of individual users. In this models, work, introducing we present the Collaborati first frame ve w Di ork rect for Preference personalized Optimization image editing (C-DPO), in diffusion a novel method that aligns image edits with user-specific preferences while leveraging collaborati as a node in ve a signals dynamic from preference like-minded graph indi and viduals.

Video Large Language Models (Video-LLMs) excel at understanding videos incontext, provided they have full access to the video when answering queries. However, these models face challenges in streaming scenarios where hour-long videos must be processed online, and questions need timely responses. In this work, we propose a training-free approach compatible with standard Video-LLMs, leveraging three key concepts: 1) LLM-informed selection of visual tokens to identify those that the LLM has attended to and contributed to its understanding of each short clip. Our attention-based selection allows us to discard up to 95% of unimportant visual tokens with minimal performance loss; 2) Recurrent processing of past selected tokens to generate temporally coherent understanding of each processed clip; 3) Caption-based question answering for lightweight and accurate responses. Our method achieves state-of-the-art performance on streaming video benchmarks, striking a balance between efficiency and effectiveness.

Warning: This method may be misused for malicious purposes. Large Language Models (LLMs) are increasingly deployed in high-risk domains.

TF-CoVR Statistics We present detailed statistics on the distribution of video counts per label in TF-CoVR, which comprises a diverse set of 306 annotated sub-actions. Both distrib video utions distrib are ution plotted for the on a F log ineGym arithmic [3] and scale F to ineDiving emphasize [6] the subsets long-tailed of TF-CoVR nature, of label frequencies. In FineGym, many labels have several hundred to over a thousand associated videos, with a gradual decline across the distribution. By contrast, FineDiving exhibits a steeper drop in video count per label, primarily due to samples, its smaller preserving dataset enough size. Ne div v ersity ertheless, to support a substantial temporal number fine-gr of ained labels composed still contain video more retrieval. A logarithmic scale is used on the y-axis to highlight the steep drop in video counts per label due to the smaller dataset size.

Models (LMMs) have demonstrated their remarkable focuses attention ties, where on has the holistic success been models gi imagev as en are general-purpose to e scaling xpected and video-language fine-grained to realize multi-modal pix pix el-le understanding.

Sparse Autoencoders (SAEs) have recently gained attention as a means to improve the interpretability and steerability of Large Language Models (LLMs), both of which are essential for AI safety. In this work, we extend the application of SAEs to Vision-Language Models (VLMs), such as CLIP, and introduce a comprehensive framework for evaluating monosemanticity at the neuron-level in visual representations. To ensure that our evaluation aligns with human perception, we propose a benchmark derived from a large-scale user study. Our experimental results reveal that SAEs trained on VLMs significantly enhance the monosemanticity of individual neurons, with sparsity and wide latents being the most influential factors. Further, we demonstrate that applying SAE interventions on CLIP's vision encoder directly steers multimodal LLM outputs (e.g., LLaVA), without any modifications to the underlying language model. These findings emphasize the practicality and efficacy of SAEs as an unsupervised tool for enhancing both interpretability and control of VLMs.

We present a theoretical framework showing that popular LLM alignment methods--including RLHF and its variants--can be understood as divergence estimators between aligned (safe or preferred) and unaligned (harmful or less-preferred) distributions. This perspective explains the emergence of separation in the latent space between safe and harmful prompts after alignment. As an application of our general divergence framework, we propose KLDO, a novel KL divergence-based alignment method, and empirically validate its effectiveness. We further show that using compliance-refusal datasets, rather than standard preference-based datasets, leads to stronger separation and improved safety alignment. Finally, to quantify the separation effect, we propose a distance-based metric in the prompt representation space, which also acts as a statistically significant indicator for model safety.