🏢 Stanford University

Researchers expose widespread prompt caching in LLMs via novel timing attacks, highlighting significant privacy risks and model architecture leakage.

Language models learn effective social deduction strategies in a virtual game by using their goal to predict useful information as a dense reward signal, doubling win rates compared to standard RL.

Boosting long-form video understanding, Temporal Preference Optimization (TPO) enhances video-LLMs by leveraging preference learning. It achieves this through a self-training method using preference …

BoxingGym: A new benchmark rigorously evaluates AI agents’ ability to design experiments and discover scientific models, revealing current LLMs’ limitations and highlighting fertile research avenues.

Whisper-GPT, a hybrid audio LLM, improves music/speech generation by combining audio waveforms and text.

FiVA dataset and its adaptation framework enable unprecedented fine-grained control over visual attributes in text-to-image generation, empowering users to craft highly customized images.

SegBook: a large-scale benchmark, reveals that fine-tuning full-body CT pre-trained models significantly improves performance on various downstream medical image segmentation tasks, particularly for s…

RedPajama, two massive open-source datasets, are released for training LLMs, improving transparency and facilitating the development of high-performing open-source models.