Large Language Models (LLMs) have shown great promise in tool-making, yet existing frameworks often struggle to efficiently construct reliable toolsets and are limited to single-task settings. To address these challenges, we propose GATE (Graph-based Adaptive Tool Evolution), an adaptive framework that dynamically constructs and evolves a hierarchical graph of reusable tools across multiple scenarios. We evaluate GATE on open-ended tasks (Minecraft), agent-based tasks (TextCraft, DABench), and code generation tasks (MATH, Date, TabMWP). Our results show that GATE achieves up to 4.3x faster milestone completion in Minecraft compared to the previous SOTA, and provides an average improvement of 9.23% over existing tool-making methods in code generation tasks and 10.03% in agent tasks. GATE demonstrates the power of adaptive evolution, balancing tool quantity, complexity, and functionality while maintaining high efficiency. Code and data are available at \url{https://github.com/ayanami2003/GATE}.

X-Ray specs and invisibility cloaks are the stuff of sci-fi and fantasy, but sometimes science is just stranger than fiction. A food dye that helps give certain sodas and snacks their hallmark orange hue renders mouse skin almost completely see-through in a reversible, potentially non-toxic research method that could transform medical and scientific imaging. Because of a counterintuitive fundamental physics principle, Tartrazine, also known as Yellow 5, can temporarily turn biological tissue transparent to the naked eye, as described in a study published September 5 in the journal Science. So far, the scientists behind the new discovery have used the method to see the organs in a mouse's intact abdomen, glimpse the pulsing vessels surrounding a rodent skull, and to get an exceptionally clear view of muscle tissue through a microscope. With further safety and efficacy research, the method may spur new scientific findings, boost microscopy advances, and improve medical diagnostic strategies and treatments.

Robotics and automation offer massive accelerations for solving intractable, multivariate scientific problems such as materials discovery, but the available search spaces can be dauntingly large. Bayesian optimization (BO) has emerged as a popular sample-efficient optimization engine, thriving in tasks where no analytic form of the target function/property is known. Here we exploit expert human knowledge in the form of hypotheses to direct Bayesian searches more quickly to promising regions of chemical space. Previous methods have used underlying distributions derived from existing experimental measurements, which is unfeasible for new, unexplored scientific tasks. Also, such distributions cannot capture intricate hypotheses. Our proposed method, which we call HypBO, uses expert human hypotheses to generate an improved seed of samples. Unpromising seeds are automatically discounted, while promising seeds are used to augment the surrogate model data, thus achieving better-informed sampling. This process continues in a global versus local search fashion, organized in a bilevel optimization framework. We validate the performance of our method on a range of synthetic functions and demonstrate its practical utility on a real chemical design task where the use of expert hypotheses accelerates the search performance significantly.

Surgical tumor removal remains one of the most common procedures during cancer treatment, with about 45% of cancer patients undergoing this surgery at some point. Thanks to recent progress in imaging and biochemical technologies, surgeons are now better able to tell tumors apart from healthy tissue. Specifically, this is enabled by a technique called "fluorescence-guided surgery" (FGS). In FGS, the patient's tissue is stained with a dye that emits infrared light when irradiated with a special light source. The dye preferentially binds to the surface of tumor cells, so that its light-wave emissions provide information on the location and extent of the tumor.

Despite not having the biggest memories in the animal kingdom, it turns out that fish are more intelligent than previously thought. Researchers at Osaka Metropolitan University in Japan have discovered that the'cleaner' wrasse fish, Labroides dimidiatus, is able to recognise itself in a photo. The species is known to attack unfamiliar fish in its territory, but when presented with an image of its own face on a stranger fish's body, it did not react to it. This suggests that the cleaner wrasse clocked the image as being one of itself through just the facial features, and so does not see it as a threatening stranger. 'This study is the first to demonstrate that fish have an internal sense of self,' said lead author Dr Masanori Kohda.

Vivian Dye is a smart woman. And so it took less than six months of selling customized golf merchandise on Zazzle, the most prominent online marketplace with three Zs in its name, for her to figure out that she needed an "Elizabeth Taylor on the beach." "I call it'Elizabeth Taylor on the beach' because she used to attract men for a cousin of hers," said Dye, referring to a plotline from an old movie. Even if customers didn't fancy her Liz, a monogrammed golfball covered with rose-colored sparkles drew users to her store, where she offered them customized pickleball paddles, shower curtains, and other golf balls. "I'm the golf ball queen," says Dye, who started selling on Zazzle in 2021, soon after retiring from her job in the mortgage industry.

A new Chameleon paint which allows people to change the colour of their cars, phone cases and trainers as many times as they want has been created by MIT researchers. The reprogrammable ink changes the colour of an object after it is exposed to ultraviolet and visible light sources at different wave lengths. A team from MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL) designed the system, which has been dubbed'PhotoChromeleon'. One researcher said the tool will allow users to personalise their appearance and belongings multiple times in a variety of styles and colours. The chameleon paint was designed by a team from MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL) which has been dubbed'PhotoChromeleon' They created the ink by mixing cyan, magenta, and yellow photochromic dyes into a single sprayable solution.

Finding the best light-harvesting chemicals for use in solar cells can feel like searching for a needle in a haystack. Over the years, researchers have developed and tested thousands of different dyes and pigments to see how they absorb sunlight and convert it to electricity. Sorting through all of them requires an innovative approach. Now, thanks to a study that combines the power of supercomputing with data science and experimental methods, researchers at the U.S. Department of Energy's (DOE) Argonne National Laboratory and the University of Cambridge in England have developed a novel "design to device" approach to identify promising materials for dye-sensitized solar cells (DSSCs). DSSCs can be manufactured with low-cost, scalable techniques, allowing them to reach competitive performance-to-price ratios.

The Apple iPhone X is flying off the shelves. Two weeks before Apple's flagship handset went on sale, I sat down to a rare on-the-record session with senior Apple execs to hear about how the phone came about, why no Home Button is better and why there's no Dark Mode. There were two Apple executives in the room: Craig Federighi, SVP of Software Engineering and Alan Dye, VP of User Interface Design. Apple has presented the iPhone X as the future. So I begin by asking how they came to decide that this was the future they wanted.

When Amy Errett set out to create Madison Reed, an online shop for hair coloring products, she and her team thought a 12-question, quiz-based algorithm was all they needed to determine the right dye for each customer's hair. But from the moment the San Francisco startup launched in 2014, women began sending in their selfies--they wanted a more visually-based assessment. Madison Reed, which prides itself on customer service, was happy to have its representatives review the selfies and assist customers, but the key to success for any tech business is the ability to scale. That's why in 2016, Madison Reed began experimenting with computer vision, a technology that applies artificial intelligence to interpret and act on visual data. Errett and her team were hoping to create photo-recognition technology that could analyze women's selfies, identify their hair, determine the color, and match it to one of the company's 46 available shades.