Many practical modeling problems involve discrete data that are best represented as draws from multinomial or categorical distributions. For example, nucleotides in a DNA sequence, children's names in a given state and year, and text documents are all commonly modeled with multinomial distributions. In all of these cases, we expect some form of dependency between the draws: the nucleotide at one position in the DNA strand may depend on the preceding nucleotides, children's names are highly correlated from year to year, and topics in text may be correlated and dynamic. These dependencies are not naturally captured by the typical Dirichlet-multinomial formulation. Here, we leverage a logistic stick-breaking representation and recent innovations in P\'{o}lya-gamma augmentation to reformulate the multinomial distribution in terms of latent variables with jointly Gaussian likelihoods, enabling us to take advantage of a host of Bayesian inference techniques for Gaussian models with minimal overhead.

Enhancing reasoning capabilities remains a central focus in the LLM reasearch community. A promising direction involves requiring models to simulate code execution step-by-step to derive outputs for given inputs. However, as code is often designed for large-scale systems, direct application leads to over-reliance on complex data structures and algorithms, even for simple cases, resulting in overfitting to algorithmic patterns rather than core reasoning structures. To address this, we propose TeaR, which aims at teaching LLMs to reason better. TeaR leverages careful data curation and reinforcement learning to guide models in discovering optimal reasoning paths through code-related tasks, thereby improving general reasoning abilities. We conduct extensive experiments using two base models and three long-CoT distillation models, with model sizes ranging from 1.5 billion to 32 billion parameters, and across 17 benchmarks spanning Math, Knowledge, Code, and Logical Reasoning. The results consistently show significant performance improvements. Notably, TeaR achieves a 35.9% improvement on Qwen2.5-7B and 5.9% on R1-Distilled-7B.

A large-scale Russian drone attack in the southeastern Ukrainian city of Dnipro has killed three people, including a young girl, according to the regional governor, hours before officials from the United States, Europe and Ukraine gather in Paris to discuss the conflict. Dnipropetrovsk Governor Serhiy Lysak said the attacks that also injured tens of people came late on Wednesday, triggering multiple fires and damaging a dozen apartment buildings. A student residence, an educational institution and a food processing plant were also damaged, Dnipro Mayor Borys Filatov added. Photos posted online showed raging fires, burned-out vehicles and buildings with shattered windows and scorched facades, as emergency crews worked through the night. Sixteen of the injured are in hospital, one of them in critical condition, according to Lysak.

Megan Gracia says her teenage son, Sewell Setzer III, died by suicide after developing a harmful attachment to an AI companion chatbot. She has filed a lawsuit against Character.AI, accusing the company of negligence. In this episode of Now You Know, Megan looks back on some of the warning signs other parents might find useful as they navigate this digital age. We also speak to one of her lawyers, Meetali Jain, about this unique case.

ChatGPT might hold a special place in your heart if it's the first AI tool you tried, but it's far from the best option now. And we aren't talking about other models like Gemini or Llama, but this other tool that combines them all in one place and eliminates the need for recurring fees: 1min.AI. It's hard to imagine how this is even possible or why everyone doesn't use 1min.AI--and that's exactly why we're sharing it. You can get a 1min.AI lifetime subscription here for 99.99, and you won't find a better price anywhere else (reg. With ChatGPT, you're used to getting only GPT models and communicating through a single text box.

We present a novel mission-planning strategy for heterogeneous multi-robot teams, taking into account the specific constraints and capabilities of each robot. Our approach employs hierarchical trees to systematically break down complex missions into manageable sub-tasks. We develop specialized APIs and tools, which are utilized by Large Language Models (LLMs) to efficiently construct these hierarchical trees. Once the hierarchical tree is generated, it is further decomposed to create optimized schedules for each robot, ensuring adherence to their individual constraints and capabilities. We demonstrate the effectiveness of our framework through detailed examples covering a wide range of missions, showcasing its flexibility and scalability.

'Special Report' host Bret Baier looks back on the evolution of media technology in covering inaugurations dating back to George Washington. At its best, today's tech makes life easier. The trick is, you need to know the insider secrets. Enter here, no purchase necessary! Here's one to make your AI results better.

Frank Cottrell-Boyce's programme, broadcast on Christmas Eve, will look at the role reading can play in shaping children's lives and find out at what happens to a child's brain when they are read to. Boxing Day will see Dwayne Fields edit a programme with items on homelessness and the value to communities of volunteering. Sir Sajid Javid's programme on 27 December will include items on the benefits and risks of artificial intelligence (AI) as well as the BBC children's programme Grange Hill. Professor Irene Tracey's programme on 28 December will see her explore developments in chronic pain relief, as well as the role that universities play in society. The quality of children's television, including an examination of the move from public service broadcasters to unregulated online platforms, will be among the topics explored on 30 December by Baroness Floella Benjamin.

Applying the word2vec technique, commonly used in language modeling, to melodies--where notes are treated as words in sentences--enables the capture of pitch information. This study examines two datasets: 20 children's songs and an excerpt from a Bach sonata. The semantic space for defining the embeddings is of very small dimension, specifically 2. Notes are predicted based on the 2, 3 or 4 preceding notes that establish the context. A multivariate analysis of the results shows that the semantic vectors representing the notes have a multiple correlation coefficient of approximately 0.80 with their pitches. Keywords Embedding; Machine Learning; Semantic meaning; Correlation 1. Introduction What kind of meaning can we capture from musical notes using word embedding techniques typically applied in language models? This study addresses this question by modeling various types of music with a relatively simple neural network, commonly used for word embedding. An embedding is a vector representation of an entity (a word, an image, a sound) in a multidimensional space where geometric relationships between vectors reflect semantic relationships between the corresponding entities (Chollet, 2021). This inquiry is not new; numerous statistical and computational models, including neural networks, have been proposed to capture key features of musical pieces and to model music perception. In 2016, Madjiheurem, Qu and Walder compared different embedding techniques to learn musical chord embeddings.