Papers

From the chassis of our cars to the skeletons of our skyscrapers, steel is the unsung hero of the modern world. But not all steel is created equal. The quest for stronger yet more ductile materials has driven the development of Advanced High-Strength Steels (AHSSs). The secret to their remarkable performance often lies in a microscopic, lightning-fast rearrangement known as the Martensitic Transformation (MT). This process is like a rapid, disciplined dance of atoms, reshaping from one crystal structure to another without diffusion or change in composition. The result is martensite—an exceptionally hard and strong phase that strengthens many AHSSs. For materials scientists, controlling this transformation is the key to designing next-generation alloys. ...

For over 50 years, scientists have been grappling with one of the grandest challenges in biology: the protein folding problem. Proteins are the microscopic workhorses of life, responsible for everything from digesting your food to fighting off viruses. Their function is dictated by their intricate three-dimensional shapes. The challenge? To predict this 3D structure solely from a protein’s one-dimensional sequence of amino acids. Solving this would be revolutionary. While billions of protein sequences have been catalogued, determining their structures experimentally—using techniques like X-ray crystallography or cryo-electron microscopy—requires painstaking work that can take months or even years. This has created a vast “structure gap” in our biological knowledge. ...

In the world of machine learning, classification is a fundamental task. From identifying spam emails to diagnosing diseases from medical images, we constantly seek to teach computers how to distinguish between categories. Over the decades, researchers have developed a range of algorithms for this—from simple linear models to multilayered neural networks. In 1995, Corinna Cortes and Vladimir Vapnik introduced a new and powerful approach: the Support-Vector Network (now widely known as the Support-Vector Machine, or SVM). Their work did more than add another tool to the classification toolbox—it brought together geometric insight and mathematical rigor into a method that remains highly influential today. ...

In the history of artificial intelligence, a handful of key ideas have acted as catalysts—transforming the field and opening up entirely new possibilities. In 1986, one such breakthrough arrived in the form of “Learning representations by back-propagating errors” by David Rumelhart, Geoffrey Hinton, and Ronald Williams. This seminal paper introduced a powerful and elegant algorithm that taught neural networks how to learn from their mistakes: backpropagation. Before this, training neural networks with multiple layers was a notoriously difficult problem. Researchers knew that deep networks should be more powerful than shallow ones, but they lacked a practical, general-purpose method to train them. Algorithms like the perceptron-convergence procedure could only handle linearly separable problems, while multilayer networks with hidden units had no effective way to adjust their weights—especially when the desired outputs of those hidden units were unknown. ...

In the history of deep learning, certain papers mark a turning point—a moment when a seemingly simple idea unlocks a new level of performance and understanding. The 2011 paper “Deep Sparse Rectifier Neural Networks” by Xavier Glorot, Antoine Bordes, and Yoshua Bengio is one such work. Before this paper, training deep neural networks was a notoriously tricky process, often requiring complex, multi-stage unsupervised pre-training to achieve good results. The standard activation functions of the time—the logistic sigmoid and hyperbolic tangent (tanh)—suffered from a critical flaw: the vanishing gradient problem. As error signals propagated backward through many layers, the gradients would shrink exponentially, making it nearly impossible for the early layers of the network to learn. ...

Large Language Models (LLMs) are evolving from simple chatbots into sophisticated agents that can use tools to accomplish complex tasks. One of the most critical tools in an agent’s toolkit is the ability to browse the web—a gateway to the world’s information. While commercial models like OpenAI’s GPT-4 and Google’s Gemini have made impressive strides, their strategies for web-browsing remain proprietary. In contrast, many open-source web agents struggle to match this performance, particularly on tasks that require deep, multi-step research. ...

Large Language Models (LLMs) have become exceptionally good at tackling tasks with clear, verifiable answers. Ask a model to solve a math problem or write a piece of code, and it often produces a correct solution by “thinking” through the problem step-by-step. This paradigm, known as deep reasoning, involves investing additional computational effort at inference time to solve complex, logical challenges—a huge reason behind the exceptional gains we’ve seen in areas like math and programming. ...

Large language models (LLMs) have reshaped what we expect from natural-language systems. Still, getting them to solve multi-step problems reliably is hard. Chain-of-Thought (CoT) prompting — asking a model to “think step-by-step” and write down its internal chain — dramatically improved performance on many tasks, from grade-school math to complex planning. But forcing models to narrate every intermediate step is costly: decoding long sequences is slow, expensive, and sometimes brittle. ...

The world of Large Language Models (LLMs) presents a classic trade-off: performance versus cost. On one hand, you have state-of-the-art models such as GPT-4 that excel at complex reasoning, sophisticated coding, and nuanced conversation. On the other, there are smaller, open-source models that are dramatically cheaper to run but often falter on demanding tasks. For real-world applications—from customer service chatbots to data analysis tools—the challenge is the same: How do you achieve the best possible performance without breaking the budget? ...

Vector embeddings have transformed information retrieval. From powering Google Search to surfacing the perfect product on Amazon, dense vector representations are now the backbone of modern search systems. We’ve asked them to tackle increasingly complex tasks—follow intricate instructions, perform multi-modal search, reason about code—and have assumed that with bigger models and better data, a single embedding vector could eventually represent any query and retrieve documents based on any notion of relevance. ...