A machine learning model is traditionally considered robust if its prediction remains (almost) constant under input perturbations with small norm. However, real-world tasks like molecular property prediction or point cloud segmentation have inherent equivariances, such as rotation or permutation equivariance. In such tasks, even perturbations with large norm do not necessarily change an input's semantic content. Furthermore, there are perturbations for which a model's prediction explicitly needs to change. For the first time, we propose a sound notion of adversarial robustness that accounts for task equivariance.

What would happen if Yellowstone's'supervolcano' erupted today? Say goodbye to Montana, Wyoming, and Idaho. More information Adding us as a Preferred Source in Google by using this link indicates that you would like to see more of our content in Google News results. This photo of a volcano in Iceland doesn't even begin to encapsulate the devastation that would happen if the Yellowstone volcano erupted. Breakthroughs, discoveries, and DIY tips sent six days a week.

Send the pool guy packing. One of these robotic buddies can maintain your water quality instead. Cleaning swimming pools is not fun. I learned this simple logic as a kid growing up in and around pools--it's the only way to survive summer in Houston, Texas. Four years ago, I became a pool owner myself, and I found that the rule still holds. Jumping into the pool on a hot day remains a rare treat, but if the pool is filled with leaves and dirt, that treat becomes a lot less delightful. And when the thermometer is reading over 100 degrees Fahrenheit, the thought of laboring on the pool deck, scooping out debris with a net, is downright cruel.

Will fusion power get cheap? New research suggests that cost declines could be slow for the technology. Fusion power could provide a steady, zero-emissions source of electricity in the future--if companies can get plants built and running. But a new study suggests that even if that future arrives, it might not come cheap. Technologies tend to get less expensive over time. Lithium-ion batteries are now about 90% cheaper than they were in 2013.

Gear 2026 Mother's Day gift guide: An updating list of great presents to give your mom No matter what your mom is into, we have perfect gift suggestions from budget-friendly options to full-on splurges. More information Adding us as a Preferred Source in Google by using this link indicates that you would like to see more of our content in Google News results. Treat your mom to the treats she deserves. We may earn revenue from the products available on this page and participate in affiliate programs. Type "gifts for mom" into a search bar and the algorithm pulls you toward the same black hole of spa baskets, quippy wine glasses, and bouquets that spew petals all over the floor.

Inference-time LLM alignment methods, particularly activation steering, offer an alternative to fine-tuning by directly modifying activations during generation. Existing methods, however, often rely on non-anticipative interventions that ignore how perturbations propagate through transformer layers and lack online error feedback, resulting in suboptimal, open-loop control. To address this, we show empirically that, despite the nonlinear structure of transformer blocks, layer-wise dynamics across multiple LLM architectures and scales are well-approximated by locally-linear models. Exploiting this property, we model LLM inference as a linear time-varying dynamical system and adapt the classical linear quadratic regulator to compute feedback controllers using layer-wise Jacobians, steering activations toward desired semantic setpoints in closed-loop with minimal computational overhead and no offline training. We also derive theoretical bounds on setpoint tracking error, enabling formal guarantees on steering performance. Using a novel adaptive semantic feature setpoint signal, our method yields robust, fine-grained behavior control across models, scales, and tasks, including state-of-the-art modulation of toxicity, truthfulness, refusal, and arbitrary concepts, surpassing baseline steering methods. Our code is available at: https://github.com/trustworthyrobotics/lqr-activation-steering

We study finite-horizon continuous-time policy evaluation from discrete closed-loop trajectories under time-inhomogeneous dynamics. The target value surface solves a backward parabolic equation, but the Bellman baseline obtained from one-step recursion is only first-order in the grid width. We estimate the time-dependent generator from multi-step transitions using moment-matching coefficients that cancel lower-order truncation terms, and combine the resulting surrogate with backward regression. The main theory gives an end-to-end decomposition into generator misspecification, projection error, pooling bias, finite-sample error, and start-up error, together with a decision-frequency regime map explaining when higher-order gains should be visible. Across calibration studies, four-scale benchmarks, feature and start-up ablations, and gain-mismatch stress tests, the second-order estimator consistently improves on the Bellman baseline and remains stable in the regime where the theory predicts visible gains. These results position high-order generator regression as an interpretable continuous-time policy-evaluation method with a clear operating region.

In biology, defects are generally bad. But in materials science, defects can be intentionally tuned to give materials useful new properties. Today, atomic-scale defects are carefully introduced during the manufacturing process of products like steel, semiconductors, and solar cells to help improve strength, control electrical conductivity, optimize performance, and more. But even as defects have become a powerful tool, accurately measuring different types of defects and their concentrations in finished products has been challenging, especially without cutting open or damaging the final material. Without knowing what defects are in their materials, engineers risk making products that perform poorly or have unintended properties.

What I've learned from 25 years of automated science, and what the future holds: an interview with Ross King We're excited to launch our new series, where we're speaking with leading researchers to explore the breakthroughs driving AI and the reality of the future promises - to give you an inside perspective on the headlines. Our first interviewee is Ross King, who created the first robot scientist back in 2009. He spoke to us about the nature of scientific discovery, the role AI has to play, and his recent work in DNA computing. Automated science is a really exciting area, and it feels like everyone's talking about it at the moment - e.g. But you've been working in this field for many years now. In 2009 you developed Adam, the first robot scientist to generate novel scientific knowledge. Could you tell me some more about that? So the history goes back to before Adam.