The SISMs of the "Tree" category (in MSRC) produced by previous SOD

The Cannes film festival isn't typically associated with video games, but this year it's playing host to an unusual collaboration. Lili is a co-production between the New York-based game studio iNK Stories (creator of 1979 Revolution: Black Friday, about a photojournalist in Iran) and the Royal Shakespeare Company, and it's been turning heads with its eye-catching translocation of Macbeth to modern-day Iran. "It's been such an incredible coup to have it as the first video game experience at Cannes," says iNK Stories co-founder Vassiliki Khonsari. "People have gone in saying, I'm not familiar playing games, so I may just try it out for five minutes. The Cannes festival's Immersive Competition began in 2024, although the lineup doesn't usually feature traditional video games. "VR films and projection mapping is the thrust of it," says iNK Stories' other co-founder, Vassiliki's husband Navid Khonsari. But Lili weaves live-action footage with video game mechanics in a similar way to a game such as Telling Lies or Immortality. Its lead, Zar Amir Ebrahimi, won best actress at Cannes three years ago. Lili focuses on the story of Lady Macbeth, here cast as the ambitious wife of an upwardly mobile officer in the Basij (a paramilitary volunteer militia within the Islamic Revolutionary Guard in Iran). As in the play, she plots a murder to secure her husband's rise. "I think that the narrative of Lady Macbeth is that she's manipulative, and that's exactly what got us interested," says Navid. "The social limitations based on her gender forced her to try to attain whatever leadership role she can," he continues. "If she was a man, she would have been one of the greatest kings that country would have ever experienced, but because she was a woman she had to work within the structure that was there for her.

One key caveat here is that we don't know much about "closed source" models--for these, companies hold back the details of how they work. Instead, we worked with researchers who measured the energy it takes to run open-source AI models, for which the source code is publicly available. But using open-source models, it's possible to directly measure the energy used to respond to a query rather than just guess. We worked with researchers who generated text, images, and video and measured the energy required for the chips the models are based on to perform the task. Even just within the text responses, there was a pretty large range of energy needs.

Breakthroughs, discoveries, and DIY tips sent every weekday. Just as birth rates decline in many wealthy and developed nations, dog parenting is remaining steady and even gaining in popularity. Up to half of households in Europe and 66 percent of homes in the United States have at least one dog and these pets are often regarded as a family member or "fur baby." To dig into what this shift says about our society, researchers from Eötvös Loránd University in Budapest, Hungary conducted a literature review to analyze the data. They propose that while dogs do not replace children, they can offer a chance to fulfill an innate nurturing drive similar to parenting, but with fewer demands than raising biological children.

R4: "Comparison to metric learning methods": We tested N-pairs [2] in our framework with a batch size of 6144.

Large scale training requires massive parallelism to finish the training within a reasonable amount of time. To support massive parallelism, large batch training is the key enabler but often at the cost of generalization performance. Existing works explore adaptive batching or hand-tuned static large batching, in order to strike a balance between the computational efficiency and the performance. However, these methods can provide only coarse-grained adaption (e.g., at a epoch level) due to the intrinsic expensive calculation or hand tuning requirements. In this paper, we propose a fully automated and lightweight adaptive batching methodology to enable fine-grained batch size adaption (e.g., at a mini-batch level) that can achieve stateof-the-art performance with record breaking batch sizes. The core component of our method is a lightweight yet efficient representation of the critical gradient noise information. We open-source the proposed methodology by providing a plugin tool that supports mainstream machine learning frameworks. Extensive evaluations on popular benchmarks (e.g., CIFAR10, ImageNet, and BERT-Large) demonstrate that the proposed methodology outperforms state-of-the-art methodologies using adaptive batching approaches or hand-tuned static strategies in both performance and batch size. Particularly, we achieve a new state-of-the-art batch size of 78k in BERT-Large pretraining with SQuAD score 90.69 compared to 90.58 reported in previous state-of-the-art with 59k batch size.

Large language models (LLMs) have made significant progress in natural language processing tasks and demonstrate considerable potential in the legal domain. However, legal applications demand high standards of accuracy, reliability, and fairness. Applying existing LLMs to legal systems without careful evaluation of their potential and limitations could pose significant risks in legal practice. To this end, we introduce a standardized comprehensive Chinese legal benchmark LexEval. This benchmark is notable in the following three aspects: (1) Ability Modeling: We propose a new taxonomy of legal cognitive abilities to organize different tasks.

Swap regret, a generic performance measure of online decision-making algorithms, plays an important role in the theory of repeated games, along with a close connection to correlated equilibria in strategic games. This paper shows an (p TN log N)-lower bound for swap regret, where T and N denote the numbers of time steps and available actions, respectively. Our lower bound is tight up to a constant, and resolves an open problem mentioned, e.g., in the book by Nisan et al. [28]. Besides, we present a computationally efficient reduction method that converts no-external-regret algorithms to no-swap-regret algorithms. This method can be applied not only to the full-information setting but also to the bandit setting and provides a better regret bound than previous results.

Rnnpool: efficient non-linear pooling for ram constrained inference.

We develop nested variational inference (NVI), a family of methods that learn proposals for nested importance samplers by minimizing an forward or reverse KL divergence at each level of nesting. NVI is applicable to many commonly-used importance sampling strategies and provides a mechanism for learning intermediate densities, which can serve as heuristics to guide the sampler. Our experiments apply NVI to (a) sample from a multimodal distribution using a learned annealing path (b) learn heuristics that approximate the likelihood of future observations in a hidden Markov model and (c) to perform amortized inference in hierarchical deep generative models. We observe that optimizing nested objectives leads to improved sample quality in terms of log average weight and effective sample size.