Transformer-based scientific foundation models are increasingly deployed in high-stakes settings, but current architectures give deterministic outputs and provide limited support for calibrated predictive uncertainty. We propose Stochastic Attention, a lightweight inference-time modification that randomizes attention by replacing softmax weights with normalized multinomial samples controlled by a single concentration parameter, and produces predictive ensembles without retraining. To set this parameter, we introduce a calibration objective that matches the stochastic attention output with the target, yielding an efficient univariate post-hoc tuning problem. We evaluate this mechanism on two scientific foundation models for weather and timeseries forecasting along with an additional regression task. Across benchmarks against uncertainty-aware baselines, we find that Stochastic Attention achieves the strongest native calibration and the sharpest prediction intervals at comparable coverage, while requiring only minutes of post-hoc tuning versus days of retraining for competitive baselines.

Neural Information Processing Systems http://nips.cc/

Supervised learning with large scale labelled datasets and deep layered models has caused a paradigm shift in diverse areas in learning and recognition. However, this approach still suffers from generalization issues under the presence of a domain shift between the training and the test data distribution. Since unsupervised domain adaptation algorithms directly address this domain shift problem between a labelled source dataset and an unlabelled target dataset, recent papers [11, 33] have shown promising results by fine-tuning the networks with domain adaptation loss functions which try to align the mismatch between the training and testing data distributions. Nevertheless, these recent deep learning based domain adaptation approaches still suffer from issues such as high sensitivity to the gradient reversal hyperparameters [11] and overfitting during the fine-tuning stage. In this paper, we propose a unified deep learning framework where the representation, cross domain transformation, and target label inference are all jointly optimized in an end-to-end fashion for unsupervised domain adaptation. Our experiments show that the proposed method significantly outperforms state-of-the-art algorithms in both object recognition and digit classification experiments by a large margin.

Neural Information Processing Systems http://nips.cc/

Neural Information Processing Systems http://nips.cc/

Life My Dad Used to Have All the Answers. It feels like he adopted a robot child, and that child will stop at nothing to wedge a divide between us. Like many twentysomethings, I ask my dad a lot of questions. How do I fix the leak under my sink? What does "federal withholding" mean?

OpenAI is facing a criminal investigation in the US over whether its ChatGPT technology played a part in the murder of two people during a mass shooting at Florida State University last year. Florida's Attorney General James Uthmeier said on Tuesday his office had been looking into the use of the artificial intelligence (AI) chatbot by a man who allegedly shot several people at the campus in Tallahassee. Our review has revealed that a criminal investigation is necessary, Uthmeier said. ChatGPT offered significant advice to this shooter before he committed such heinous crimes. An OpenAI spokesperson said: ChatGPT is not responsible for this terrible crime.

ChatGPT has been connected to at least two mass shootings in the last year. Florida Attorney General James Ulthmeier has announced that the state's Office of Statewide Prosecution has opened a criminal investigation into OpenAI and ChatGPT. The investigation was opened because the suspect in a mass shooting at Florida State University in 2025 reportedly used ChatGPT in the lead up to the shooting. Per Uthmeier, Florida law states that anyone who aids, abets, or counsels someone in the commission of a crime, and that crime is committed or attempted, may be considered a principal to the crime. That means that the responses provided by ChatGPT to the shooter could be interpreted as the AI assistant aiding and abetting his actions.

The ChatGPT Images 2.0 model is here. Our testing shows it's better at creating more detailed images and rendering text, but it still struggles with languages other than English. OpenAI launched a new image generation AI model on Tuesday, dubbed ChatGPT Images 2.0. This model can generate more than one image from a single prompt, like an entire study booklet, as well as output text, including in non-English languages, like Chinese and Hindi. This release is available globally for ChatGPT and Codex users, with a more powerful version available for paying subscribers.

ChatGPT Images 2.0 is better at rendering non-Latin text OpenAI describes it as a step change for image generation models. OpenAI's new ChatGPT Images 2.0 model is now available. A little more than a year after OpenAI gave ChatGPT users the option to create images and designs directly from its chatbot, it's now releasing ChatGPT Images 2.0 . OpenAI describes the new system as a "step change" for image generation models, particularly when it comes to the tool's ability to follow instructions in detail, render dense text and place and relate objects in a scene. For the first time, OpenAI has also built an image model with reasoning capabilities, giving the system the ability to do things like search the web and verify its outputs.