Uncategorized

With the advancements in Natural Language Processing and Natural Language Generation, Large Language Models (LLMs) are being frequently used in real-world applications. With their ability to mimic human behavior, these models, with their general-purpose nature, have stepped into every field and domain.  Though these models have gained significant attention, these models represent a constrained and…

In machine learning, predicting outcomes accurately is crucial, but it’s equally important to understand the uncertainty associated with those predictions. Uncertainty helps us gauge our confidence in a model’s output. However, not all machine learning models provide this uncertainty information. This can lead to situations where decisions are made based on overly optimistic predictions, potentially…

Artificial intelligence is advancing rapidly, but researchers are facing a significant challenge. AI systems struggle to adapt to diverse environments outside their training data, which is critical in areas like self-driving cars, where failures can have catastrophic consequences. Despite efforts by researchers to tackle this problem with algorithms for domain generalization, no algorithm has yet…

Protein engineering, a field with wide-ranging applications in chemistry, energy, and medicine, has multiple intricate challenges. Existing methods of engineering new proteins with improved or novel functions are slow, labor-intensive, and inefficient. This inefficiency in protein engineering hampers the ability to exploit its potential in various scientific and medical fields. Protein engineering involves a discovery-driven…

Artificial Intelligence (AI) has made significant strides in the past few years with the advancements in Deep Learning (DL) and the advent of Large Language Models (LLMs). Many powerful applications have been developed that are capable of processing enormous amounts of data. Although these innovations speed up and optimize many aspects of our work, they…

Diffusion models are at the forefront of generative model research. These models, essential in replicating complex data distributions, have shown remarkable success in various applications, notably in generating intricate and realistic images. They establish a stochastic process that progressively adds noise to data, followed by a learned reversal of this process to create new data…

The landscape of language models is evolving rapidly, driven by the empirical success of scaling models with increased parameters and computational budgets. In this era of large language models, Mixture-of-Experts (MoE) architecture emerges as a key player, offering a solution to manage computational costs while scaling model parameters. However, challenges persist in ensuring expert specialization…

Molecular property prediction stands at the forefront of drug discovery and design, which has grown increasingly dependent on advancements in artificial intelligence and machine learning. Traditional methods, while foundational, often need to catch up in their scope, unable to encapsulate the vast and intricate details of molecular characteristics. This gap in capability highlights the need…

The Billion-Scale Approximate Nearest Neighbor Search Challenge, part of the NeurIPS competition track, aims to advance research in large-scale ANNS (Approximate Nearest Neighbor Search).  BigANN is a collaborative arena where the best minds in the field come together to push the boundaries of vector search technology. Participants face four distinct tracks, each tackling a different…

MLLMs, or multimodal large language models, have been advancing lately. By incorporating images into large language models (LLMs) and harnessing the capabilities of LLMs, MLLMs demonstrate exceptional skill in tasks including visual question answering, instruction following, and image understanding. Studies have seen a significant flaw in these models despite their improvements; they still have some…