A new AI tool can guess what you're looking at just by analyzing your brain scans--and recreate that image with remarkable precision. It can go the other way too, and predict a person's brain activity based on what they're looking at. In the image above, for example, the left-hand image of each pair is what the user actually saw--and its right-hand counterpart is what the model recreated based on the brain scan. Michal Irani, who developed the tool with her colleagues at the Weizmann Institute of Science in Rehovot, Israel, hopes her "mindreading" tool will ultimately reveal more about how the brain works, and could perhaps be used to help locked-in people communicate, or allow scientists to recreate the content of dreams. Judy Illes, a neuroethicist and professor of neurology at the University of British Columbia in Canada, who was not involved in the research, describes the work as "magnificent." "The idea [of using this approach] to help people with neurologic conditions therapeutically is tremendously exciting," she says. But other scientists warn that a similar approach could be used to reveal the inner thoughts and mental imagery of people, potentially without their consent. "The results seem very impressive," says Tommy Sprague, a neuroscientist at the University of California Santa Barbara. "But if there's a way to surreptitiously extract information about what you're thinking about, then 150 years of sci-fi can come true anytime, and that's worrisome in a lot of ways."

Overlap, also known as positivity, is a key condition for causal treatment effect estimation. Many popular estimators suffer from high variance and become brittle when features differ strongly across treatment groups. This is especially challenging in high dimensions: the curse of dimensionality can make overlap implausible. To address this, we propose a class of feature representations called deconfounding scores, which preserve both identification and the target of estimation; the classical propensity and prognostic scores are two special cases. We characterize the problem of finding a representation with better overlap as minimizing an overlap divergence under a deconfounding score constraint. We then derive closed-form expressions for a class of deconfounding scores under a broad family of generalized linear models with Gaussian features and show that prognostic scores are overlap-optimal within this class. We conduct extensive experiments to assess this behavior empirically.

Property prediction on molecular graphs is an important application of Graph Neural Networks (GNNs).

Recent studies in reinforcement learning (RL) have made significant progress by leveraging function approximation to alleviate the sample complexity hurdle for better performance. Despite the success, existing provably efficient algorithms typically rely on the accessibility of immediate feedback upon taking actions. The failure to account for the impact of delay in observations can significantly degrade the performance of real-world systems due to the regret blow-up. In this work, we tackle the challenge of delayed feedback in RL with linear function approximation by employing posterior sampling, which has been shown to empirically outperform the popular UCB algorithms in a wide range of regimes. We first introduce Delayed-PSVI, an optimistic value-based algorithm that effectively explores the value function space via noise perturbation with posterior sampling.

Advances in generative models have made it possible for AI-generated text, code, and images to mirror human-generated content in many applications. W atermark-ing, a technique that aims to embed information in the output of a model to verify its source, is useful for mitigating the misuse of such AI-generated content. However, we show that common design choices in LLM watermarking schemes make the resulting systems surprisingly susceptible to attack--leading to fundamental trade-offs in robustness, utility, and usability. To navigate these trade-offs, we rigorously study a set of simple yet effective attacks on common watermarking systems, and propose guidelines and defenses for LLM watermarking in practice.

We collect 312 programs from various sources, including daily programs from college homework, the international competition (SV -COMP), benchmarks from previous papers (SLING), and programs from real-world software systems (Linux Kernel, GlibC, LiteOS, and Zephyr).