Related papers: An efficient forgetting-aware fine-tuning framewor…
The continual learning capability of large language models (LLMs) is crucial for advancing artificial general intelligence. However, continual fine-tuning LLMs across various domains often suffers from catastrophic forgetting, characterized…
Class-incremental Learning (CIL) enables the model to incrementally absorb knowledge from new classes and build a generic classifier across all previously encountered classes. When the model optimizes with new classes, the knowledge of…
Large Language Models (LLMs) are often English-centric due to the disproportionate distribution of languages in their pre-training data. Enhancing non-English language capabilities through post-pretraining often results in catastrophic…
Machine-learned interatomic potentials (MLIPs) are increasingly used to replace computationally demanding electronic-structure calculations to model matter at the atomic scale. The most commonly used model architectures are constrained to…
To tackle the issues of catastrophic forgetting and overfitting in few-shot class-incremental learning (FSCIL), previous work has primarily concentrated on preserving the memory of old knowledge during the incremental phase. The role of…
We introduce Flexible Cutoff Learning (FCL), a method for training machine learning interatomic potentials (MLIPs) whose cutoff radii can be adjusted after training. Unlike conventional MLIPs that fix the cutoff radius during training, FCL…
Machine learned interatomic potentials (MLIPs) are becoming a standard method for DFT-level accurate molecular dynamics simulation and large-scale studies of crystal energetics. Increasingly popular are universal pre-trained potentials,…
Machine learning (ML) based interatomic potentials are emerging tools for materials simulations but require a trade-off between accuracy and speed. Here we show how one can use one ML potential model to train another: we use an existing,…
Pretrained models like CLIP have demonstrated impressive zero-shot classification capabilities across diverse visual domains, spanning natural images, artistic renderings, and abstract representations. However, real-world applications often…
Machine learning interatomic potentials (MLIPs) are changing atomistic simulations in the field of chemistry and materials science. However, constructing a single universal MLIP that can accurately model molecular and crystalline systems…
Continual learning has become essential in many practical applications such as online news summaries and product classification. The primary challenge is known as catastrophic forgetting, a phenomenon where a model inadvertently discards…
Reinforcement Learning (RL) has been successful in various domains like robotics, game playing, and simulation. While RL agents have shown impressive capabilities in their specific tasks, they insufficiently adapt to new tasks. In…
Machine learning interatomic potentials (MLIPs) have proven to be wildly useful for molecular dynamics simulations, powering countless drug and materials discovery applications. However, MLIPs face two primary bottlenecks preventing them…
Accurate yet transferable machine-learning interatomic potentials (MLIPs) are essential for accelerating materials and chemical discovery. However, most universal MLIPs overfit to narrow datasets or computational protocols, limiting their…
We propose an approach to materials prediction that uses a machine-learning interatomic potential to approximate quantum-mechanical energies and an active learning algorithm for the automatic selection of an optimal training dataset. Our…
Universal machine learning interatomic potentials (uMLIPs) deliver near ab initio accuracy in energy and force calculations at low computational cost, making them invaluable for materials modeling. Although uMLIPs are pre-trained on vast ab…
Machine-learned interatomic potentials (MLIPs) have become the gold standard for atomistic simulations, yet their extension to magnetic materials remains challenging because spin fluctuations must be captured either explicitly or…
Pre-trained models have demonstrated exceptional generalization capabilities in time-series forecasting; however, adapting them to evolving data distributions remains a significant challenge. A key hurdle lies in accessing the original…
We unveil that internal representations in large language models (LLMs) serve as reliable proxies of learned knowledge, and propose RECALL, a novel representation-aware model merging framework for continual learning without access to…
Machine learning (ML) models utilizing structure-based features provide an efficient means for accurate property predictions across diverse chemical spaces. However, obtaining equilibrium crystal structures typically requires expensive…