You're viewing the US edition You can switch to the UK or AU homepage at any time using this menu. Dolly Parton modelled her look on the'town tramp' and never denied repeated affair rumours - but her brilliant business brain meant she could have been a billionaire if she hadn't given so much to charity What the Sussexes' new life in Britain will really look like: It's Harry's dream... Meghan will'come and go' - but it's all part of the 15-year plan. Dolly Parton's heartbreaking final year: The deaths of her beloved husband and brother, devastating health struggles... and the comeback she was forced to abandon The photos that sparked wild internet rumours: James Blunt mingles with Hayden Panettiere on a yacht in 2009 after he broke his silence on claims he was the'British singer' who the late actress was put into bed with aged 18 Dolly Parton's health woes as family reveals intimate funeral plans after country legend's death at 80: Live updates Chilling 911 call details how police watched Montana home go up in flames before 8 found dead inside... as shooting suspect's haunting final act is revealed Dolly Parton's devastating final interview just FOUR DAYS before her death: 'I still have so many things I want to achieve' Fleet of'doomsday planes' spotted over US hours after CIA boss visits Moscow Lindsay Clancy claimed she heard chilling'voice' telling her to kill her three children while she made them dinner, final witness says: Live updates My wife and I are in our 50s and haven't had sex for TWO YEARS. Today I'm revealing all about the desperate truth so many couples hide... by CHRIS TAUNTON Trump's CIA chief John Ratcliffe makes surprise trip to Moscow for secret meeting amid grim fears for Poland Menopause made my hair fall out. Then I discovered a new hair-loss solution.

You're viewing the US edition You can switch to the UK or AU homepage at any time using this menu. James Blunt breaks silence on claims he was'famous British singer' who Hayden Panettiere was put into bed with when she was 18 Brian Hickerson's dark family secrets outed by his own sister... who reveals moment his'demons' took over... and real reason Hayden Panettiere kept going back His penis grew by a THIRD in minutes after game-changing new enlargement method... now thrilled patient bares all and declares: 'My phone's already buzzing' No one will dare say it out loud... but there's only one bombshell way Harry and Meghan's'half-in, half-out' new life in Britain can work: ROBERT JOBSON Kim Kardashian and Kanye West's daughter North, 13, says she has no interest in'boys, vaping and doing bad stuff' as she appears in bold Dazed cover shoot Mother of three young boys who vanished on Thanksgiving 2010 while being cared for by their dad admits she is a PEDOPHILE at ex's murder trial The 10-cent supplement that turbocharges your Mounjaro AND reduces your cholesterol: Most patients won't have heard of it, but now experts reveal why anyone on jabs must take it - and how it can stop you putting weight back on Enes Kanter Freedom hits back at'terrified' WNBA as he learns draft fate after ex-NBA star was ejected from game over explosive courtside clash Hoover Dam's horror threat to 40 million Americans: As'tipping point' is reached... expert warns residents in THREE states to brace Distressing audio reveals moment doctor's husband who left daughter, 2, to die in hot car was found by nanny after killing himself just hours before he was due to be jailed Taylor Swift reveals her'coping mechanism' for dealing with critics and admits she likes to'mock them back' Glamorous Duke University influencer accused of astonishingly sly plot... as students reveal blistering evidence Harry and Meghan want to move to the Cotswolds... but the party I attended last weekend - and these neighbourhood WhatsApp chats - show why they won't get the warm welcome they're hoping for: NADINE DORRIES Hidden in plain sight: Stone from Henry VIII's long-lost Sandown Castle was reused in Dover Castle, study finds Part of King Henry VIII's long-lost Sandown Castle has been rediscovered hiding in plain sight. Stones from the 16th-century fort were reused to build part of Dover Castle, about nine miles (14.5km) down the coast, according to a new study. Archaeologists discovered a previously unknown report from 1883, stating that materials from Sandown were being sold off and sent to Dover. Those stone chunks from Henry VIII's maritime defences now form part of the Constable's Gate at Dover, which provided quarters for the castle's commanding officer.

The medieval'cog' was nearly 92 feet long and featured castles on its bow and stern. Breakthroughs, discoveries, and DIY tips sent every weekday. Archaeologists in Denmark say a sunken Viking ship near Copenhagen is the largest boat of its kind ever discovered--and that's saying something. At nearly 92 feet long, the 600-year-old vessel is also one of the best preserved examples of a cog, a "super ship" whose advanced design and carrying capacity helped transform trade in medieval Europe. "The find is a milestone for maritime archaeology," excavation lead Otto Uldum said in a statement, adding the boat now offers a "unique opportunity to understand both the construction and life on board the biggest trading ships of the Middle Ages."

Regularization improves generalization of supervised models to out-of-sample data. Prior works have shown that prediction in the causal direction (effect from cause) results in lower testing error than the anti-causal direction. However, existing regularization methods are agnostic of causality. We introduce Causal Structure Learning (CASTLE) regularization and propose to regularize a neural network by jointly learning the causal relationships between variables. CASTLE learns the causal directed acyclical graph (DAG) as an adjacency matrix embedded in the neural network's input layers, thereby facilitating the discovery of optimal predictors. Furthermore, CASTLE efficiently reconstructs only the features in the causal DAG that have a causal neighbor, whereas reconstruction-based regularizers suboptimally reconstruct all input features. We provide a theoretical generalization bound for our approach and conduct experiments on a plethora of synthetic and real publicly available datasets demonstrating that CASTLE consistently leads to better out-of-sample predictions as compared to other popular benchmark regularizers.

Activation steering has emerged as a powerful method for guiding the behavior of generative models towards desired outcomes such as toxicity mitigation. However, most existing methods apply interventions uniformly across all inputs, degrading model performance when steering is unnecessary. We introduce Dynamically Scaled Activation Steering (DSAS), a method-agnostic steering framework that decouples when to steer from how to steer. DSAS adaptively modulates the strength of existing steering transformations across layers and inputs, intervening strongly only when undesired behavior is detected. At generation time, DSAS computes context-dependent scaling factors that selectively adjust the strength of any steering method. We also show how DSAS can be jointly optimized end-to-end together with the steering function. When combined with existing steering methods, DSAS consistently improves the Pareto front with respect to steering alone, achieving a better trade-off between toxicity mitigation and utility preservation. We further demonstrate DSAS's generality by applying it to a text-to-image diffusion model, showing how adaptive steering allows the modulation of specific concepts. Finally, DSAS introduces minimal computational overhead while improving interpretability, pinpointing which tokens require steering and by how much.

We thank all the reviewers for their valuable suggestions and feedback. Table 2 contains L2 regularization. 's, of the proposed neural network (Section 3.3). We will elaborate this point with concrete examples in the revised submission. The description of the regularization terms is given in lines 162-168.

In standard causal attention, each token's query, key, and value (QKV) are static and encode only preceding context. We introduce CAuSal aTtention with Lookahead kEys (CASTLE), an attention mechanism that continually updates each token's keys as the context unfolds. We term these updated keys lookahead keys because they belong to earlier positions yet integrate information from tokens that appear later relative to those positions, while strictly preserving the autoregressive property. Although the mechanism appears sequential, we derive a mathematical equivalence that avoids explicitly materializing lookahead keys at each position and enables efficient parallel training. On language modeling benchmarks, CASTLE consistently outperforms standard causal attention across model scales, reducing validation perplexity and improving performance on a range of downstream tasks.

We introduce a real-time strategy game environment based on Generals.io, a game with thousands of weekly active players. Our environment is fully compatible with Gymnasium and PettingZoo and is capable of running thousands of frames per second on commodity hardware. We also present a reference agent, trained with supervised pre-training and self-play, which reached the top 0.003% of the 1v1 human leaderboard after only 36 hours on a single H100 GPU. To accelerate learning, we incorporate potential-based reward shaping and memory features. Our contributions of a modular RTS benchmark and a competitive baseline agent provide an accessible yet challenging platform for advancing multi-agent reinforcement learning research. The documented code, together with examples and tutorials, is available at https://github.com/strakam/generals-bots.