Lego's next release is a digital brick loaded with sensors that add new layers of interactivity to its play sets. WIRED got exclusive access to the Lego labs where the Smart Brick was born. The secretive division of 237 staff based here and in London, Boston, and Singapore is dedicated to thinking up what comes next for the world's largest toy brand. In front of me, on a plain white table, is a batch of prototypes of Lego's new Smart Brick, the final version of which is a small, sensor-laden 2-by-4 black brick with a big brain. No outsider has seen these prototypes, all of which represent stages of a journey Lego has been charting over the past eight years. Lego hopes this innovation, which lands in stores March 1, will safeguard the future of its plastic empire. The diminutive proportions of the finished Smart Brick belie the fact that the thing is exceedingly clever. Inside is a tiny custom chip running bespoke software that can communicate with onboard sensors to monitor and react to motion, orientation, and magnetic fields. It's also likely no exaggeration that the Smart Brick could represent the most radical product Lego has produced since Jens Nygaard Knudsen, the company's former longtime chief designer, created the minifigure nearly 50 years ago.

Table 9) and identifying reflections (Error #20 in Table 13) are also noted.

The Chinese video game giant Tencent is now building some of the world's best 3D AI models. This could have implications far outside game design. The video game Valorant, a fast-paced team-based shooter, has recently become a testing ground for a promising new direction in artificial intelligence research. The game's developers at Riot Games (a Tencent subsidiary) are using 3D-native AI models to prototype new characters, scenes, and storylines, according to a researcher familiar with the company's efforts who spoke on the condition of anonymity. While many AI models can generate text, images, and video, Tencent's Hunyuan (混元 or "first mix") family of models can dream up 3D objects and interactive scenes.

Quantile Regression Forests (QRF) are widely used for non-parametric conditional quantile estimation, yet statistical inference for variable importance measures remains challenging due to the non-smoothness of the loss function and the complex bias-variance trade-off. In this paper, we develop a asymptotic theory for variable importance defined as the difference in pinball loss risks. We first establish the asymptotic normality of the QRF estimator by handling the non-differentiable pinball loss via Knight's identity. Second, we uncover a "phase transition" phenomenon governed by the subsampling rate $β$ (where $s \asymp n^β$). We prove that in the bias-dominated regime ($β\ge 1/2$), which corresponds to large subsample sizes typically favored in practice to maximize predictive accuracy, standard inference breaks down as the estimator converges to a deterministic bias constant rather than a zero-mean normal distribution. Finally, we derive the explicit analytic form of this asymptotic bias and discuss the theoretical feasibility of restoring valid inference via analytic bias correction. Our results highlight a fundamental trade-off between predictive performance and inferential validity, providing a theoretical foundation for understanding the intrinsic limitations of random forest inference in high-dimensional settings.

Modern language models fail a fundamental requirement of trustworthy intelligence: knowing when not to answer. Despite achieving impressive accuracy on benchmarks, these models produce confident hallucinations, even when wrong answers carry catastrophic consequences. Our evaluations on GSM8K, MedQA and GPQA show frontier models almost never abstain despite explicit warnings of severe penalties, suggesting that prompts cannot override training that rewards any answer over no answer. As a remedy, we propose Reinforced Hesitation (RH): a modification to Reinforcement Learning from Verifiable Rewards (RLVR) to use ternary rewards (+1 correct, 0 abstention, -$λ$ error) instead of binary. Controlled experiments on logic puzzles reveal that varying $λ$ produces distinct models along a Pareto frontier, where each training penalty yields the optimal model for its corresponding risk regime: low penalties produce aggressive answerers, high penalties conservative abstainers. We then introduce two inference strategies that exploit trained abstention as a coordination signal: cascading routes queries through models with decreasing risk tolerance, while self-cascading re-queries the same model on abstention. Both outperform majority voting with lower computational cost. These results establish abstention as a first-class training objective that transforms ``I don't know'' from failure into a coordination signal, enabling models to earn trust through calibrated honesty about their limits.

Reinforcement learning (RL) has emerged as a powerful paradigm for enhancing the reasoning capabilities of large language models (LLMs). While RL has demonstrated substantial performance gains, it still faces key challenges, including low sampling efficiency and a strong dependence on model initialization: some models achieve rapid improvements with minimal RL steps, while others require significant training data to make progress. In this work, we investigate these challenges through the lens of reasoning token coverage and argue that initializing LLMs with diverse, high-quality reasoning primitives is essential for achieving stable and sample-efficient RL training. We propose Tailor, a finetuning pipeline that automatically discovers and curates novel reasoning primitives, thereby expanding the coverage of reasoning-state distributions before RL. Extensive experiments on mathematical and logical reasoning benchmarks demonstrate that Tailor generates more diverse and higher-quality warm-start data, resulting in higher downstream RL performance.

AgiBot is using AI-powered robots to do new manufacturing tasks. Smarter machines may transform physical labor in China. AgiBot, a humanoid robotics company based in Shanghai, has engineered a way for two-armed robots to learn manufacturing tasks through human training and real-world practice on a factory production line. The company says its system, which combines teleoperation and reinforcement learning, is being tested on a production line belonging to Longcheer Technology, a Chinese company that manufactures smartphones, VR headsets, and other electronic gadgets. AgiBot's project shows how more advanced AI is starting to change the abilities of industrial machines--an innovation that may creep into new areas of manufacturing in China and elsewhere.

Contemporary chess engines offer precise yet opaque evaluations, typically expressed as centipawn scores. While effective for decision-making, these outputs obscure the underlying contributions of individual pieces or patterns. In this paper, we explore adapting SHAP (SHapley Additive exPlanations) to the domain of chess analysis, aiming to attribute a chess engines evaluation to specific pieces on the board. By treating pieces as features and systematically ablating them, we compute additive, per-piece contributions that explain the engines output in a locally faithful and human-interpretable manner. This method draws inspiration from classical chess pedagogy, where players assess positions by mentally removing pieces, and grounds it in modern explainable AI techniques. Our approach opens new possibilities for visualization, human training, and engine comparison. We release accompanying code and data to foster future research in interpretable chess AI.

A new study shows that feeding large language models low-quality, high-engagement content from social media lowers their cognitive abilities. AI models may be a bit like humans, after all. A new study from the University of Texas at Austin, Texas A&M, and Purdue University shows that large language models fed a diet of popular but low-quality social media content experience a kind of "brain rot" that may be familiar to anyone who has spent too long doomscrolling on X or TikTok. We live in an age where information grows faster than attention spans--and much of it is engineered to capture clicks, not convey truth or depth," says Junyuan Hong, an incoming assistant professor at the National University of Singapore who worked on the study as a graduate student at UT Austin. "We wondered: What happens when AIs are trained on the same stuff?"

Abstract--Spiking Neural Networks are attracting increased attention as a more energy-efficient alternative to traditional Artificial Neural Networks for edge computing. Neuromorphic computing can significantly reduce energy requirements. Here, we present a complete pipeline: efficient event-based training of SNNs with synaptic delays on GPUs and deployment on Intel's Loihi 2 neuromorphic chip. We evaluate our approach on keyword recognition tasks using the Spiking Heidelberg Digits and Spiking Speech Commands datasets, demonstrating that our algorithm can enhance classification accuracy compared to architectures without delays. Our benchmarking indicates almost no accuracy loss between GPU and Loihi 2 implementations, while classification on Loihi 2 is up to 18 faster and uses 250 less energy than on an NVIDIA Jetson Orin Nano.