There's a surprisingly competitive global race on to grow a 3,000-pound pumpkin. Ian (left) and Stuart Paton pose with a giant pumpkin in their nursery in the New Forest, Hampshire. Breakthroughs, discoveries, and DIY tips sent every weekday. The pumpkin's name was Muggle and it weighed as much as a bull moose. At 2,819 pounds and over 21 feet in circumference, this enormous gourd claimed the dual titles of "heaviest pumpkin" and "largest pumpkin by circumference" in the on October 4, 2025.

This paper investigates the identification of the top-m user-scheduling sets in multi-user MIMO downlink, which is cast as a combinatorial pure-exploration problem in stochastic linear bandits. Because the action space grows exponentially, exhaustive search is infeasible. We therefore adopt a linear utility model to enable efficient exploration and reliable selection of promising user subsets. We introduce a gap-index framework that maintains a shortlist of current estimates of champion arms (top-m sets) and a rotating shortlist of challenger arms that pose the greatest threat to the champions. This design focuses on measurements that yield the most informative gap-index-based comparisons, resulting in significant reductions in runtime and computation compared to state-of-the-art linear bandit methods, with high identification accuracy. The method also exposes a tunable trade-off between speed and accuracy. Simulations on a realistic OFDM downlink show that shortlist-driven pure exploration makes online, measurement-efficient subcarrier selection practical for AI-enabled communication systems.

The Busy Beaver Challenge, a notoriously difficult question in theoretical computer science, is now producing answers so large they're impossible to write out using standard mathematical notation. Imagine that someone gives you a list of five numbers: 1, 6, 21, 107 and--wait for it--47,176,870. Can you guess what comes next? These are the first five busy beaver numbers. They form a sequence that's intimately tied to one of the most notoriously difficult questions in theoretical computer science. Determining the values of busy beaver numbers is a daunting challenge that has attracted a cult following among both professional and amateur mathematicians for over 60 years. Researchers identified the first four busy beaver numbers in the 1960s and 1970s.

Recent advances in long-term time series forecasting have introduced numerous complex prediction models that consistently outperform previously published architectures. However, this rapid progression raises concerns regarding inconsistent benchmarking and reporting practices, which may undermine the reliability of these comparisons. Our position emphasizes the need to shift focus away from pursuing ever-more complex models and towards enhancing benchmarking practices through rigorous and standardized evaluation methods. To support our claim, we first perform a broad, thorough, and reproducible evaluation of the top-performing models on the most popular benchmark by training 3,500+ networks over 14 datasets. Then, through a comprehensive analysis, we find that slight changes to experimental setups or current evaluation metrics drastically shift the common belief that newly published results are advancing the state of the art. Our findings suggest the need for rigorous and standardized evaluation methods that enable more substantiated claims, including reproducible hyperparameter setups and statistical testing.

When a Finn talks to an AI helper like ChatGPT, they often get the sense that something is subtly wrong. "You really feel that this conversation is not the way that you would have a discussion in Finland," says Peter Sarlin. For a start, Finnish people are known for a blunt approach to dialog and chatbots are usually tuned to be overly courteous. But there's also the fact that most leading chatbots and the large language models behind them are developed in the US and trained on mostly US data. Cutting-edge AI products often come with a tonality that is essentially American.

Arthur Mensch, tall and lean with a flop of unkempt hair, arrived for a speech last month at a sprawling tech hub in Paris wearing jeans and carrying a bicycle helmet. He had an unassuming look for a person European officials are counting on to help propel the region into a high-stakes match with the United States and China over artificial intelligence. Mensch, 31, is the CEO and a founder of Mistral, considered by many to be one of the most promising challengers to OpenAI and Google. "You have become the poster child for AI in France," Matt Clifford, a British investor, told him onstage. A lot is riding on Mensch, whose company has shot into the spotlight just a year after he founded it in Paris with two college friends.

The navigation of complex labyrinths with tens of rooms under visual partially observable state is typically addressed using recurrent deep reinforcement learning architectures. In this work, we show that navigation can be achieved through the emergent evolution of a simple Braitentberg-style heuristic that structures the interaction between agent and labyrinth, i.e. complex behaviour from simple heuristics. To do so, the approach of tangled program graphs is assumed in which programs cooperatively coevolve to develop a modular indexing scheme that only employs 0.8\% of the state space. We attribute this simplicity to several biases implicit in the representation, such as the use of pixel indexing as opposed to deploying a convolutional kernel or image processing operators.

Temporal knowledge graph (TKG) forecasting benchmarks challenge models to predict future facts using knowledge of past facts. In this paper, we apply large language models (LLMs) to these benchmarks using in-context learning (ICL). We investigate whether and to what extent LLMs can be used for TKG forecasting, especially without any fine-tuning or explicit modules for capturing structural and temporal information. For our experiments, we present a framework that converts relevant historical facts into prompts and generates ranked predictions using token probabilities. Surprisingly, we observe that LLMs, out-of-the-box, perform on par with state-of-the-art TKG models carefully designed and trained for TKG forecasting. Our extensive evaluation presents performances across several models and datasets with different characteristics, compares alternative heuristics for preparing contextual information, and contrasts to prominent TKG methods and simple frequency and recency baselines. We also discover that using numerical indices instead of entity/relation names, i.e., hiding semantic information, does not significantly affect the performance ($\pm$0.4\% Hit@1). This shows that prior semantic knowledge is unnecessary; instead, LLMs can leverage the existing patterns in the context to achieve such performance. Our analysis also reveals that ICL enables LLMs to learn irregular patterns from the historical context, going beyond simple predictions based on common or recent information.

Artificial intelligence (AI) in sport may sound like something from a science fiction movie. But unbelievable footage has captured the moment an AI pilot overthrows three human champions in a fierce drone race. Drone racing world champion, Alex Vanover, was among the trio of professionals thrashed by AI as it won an impressive 15 of 25 trials in a Swiss-based study. The so-called'Swift' technology mastered speeds of over 62mph (100kmh), with a reaction time that was 120 milliseconds faster than humans on average. Its impressive track record is a milestone for scientists at the University of Zurich, who sought to move this pioneering technology beyond board games.

As this year's RoboCup draws to a close, we take a look back at some of the highlights from the second half of the conference. Over the course of the weekend, the event focussed on the latter stages of the competitions, with the winners in all the different leagues decided. If you missed our round-up of the first half of RoboCup, you can catch up here. It will be a day of intense games. Time is flying when you are having fun.