相关论文: Exploring the Hierarchical Reasoning Model for Sma…

The success of deep learning in computer vision has been driven by models of increasing scale, from deep Convolutional Neural Networks (CNN) to large Vision Transformers (ViT). While effective, these architectures are parameter-intensive…

Neural networks have proven effective at solving difficult problems but designing their architectures can be challenging, even for image classification problems alone. Our goal is to minimize human participation, so we employ evolutionary…

Current large language models (LLMs) primarily rely on linear sequence generation and massive parameter counts, yet they severely struggle with complex algorithmic reasoning. While recent reasoning architectures, such as the Hierarchical…

Deep neural networks often rely on spurious features to make predictions, which makes them brittle under distribution shift and on samples where the spurious correlation does not hold (e.g., minority-group examples). Recent studies have…

Deep learning based medical image recognition systems often require a substantial amount of training data with expert annotations, which can be expensive and time-consuming to obtain. Recently, synthetic augmentation techniques have been…

Recursive reasoning models such as Hierarchical Reasoning Model (HRM) and Tiny Recursive Model (TRM) show that small, weight-shared networks can solve compute-heavy and NP puzzles by iteratively refining latent states, but their training…

The adaptation of large language models (LLMs) to specialized reasoning tasks is fundamentally constrained by computational resources. Parameter-Efficient Fine-Tuning (PEFT) methods have emerged as a powerful solution, yet the landscape of…

Few-shot learning has been studied to adapt models to tasks with very few samples. It holds profound significance, particularly in clinical tasks, due to the high annotation cost of medical images. Several works have explored few-shot…

Hierarchical reasoning model (HRM) achieves extraordinary performance on various reasoning tasks, significantly outperforming large language model-based reasoners. To understand the strengths and potential failure modes of HRM, we conduct a…

Recently, various pre-trained language models (PLMs) have been proposed to prove their impressive performances on a wide range of few-shot tasks. However, limited by the unstructured prior knowledge in PLMs, it is difficult to maintain…

This article demonstrates a new conceptor network based classifier in classifying images. Mathematical descriptions and analysis are presented. Various tests are experimented using three benchmark datasets: MNIST, CIFAR-10 and CIFAR-100.…

Machine learning training methods depend plentifully and intricately on hyperparameters, motivating automated strategies for their optimisation. Many existing algorithms restart training for each new hyperparameter choice, at considerable…

We propose Diff-Instruct* (DI*), a data-efficient post-training approach for one-step text-to-image generative models to improve its human preferences without requiring image data. Our method frames alignment as online reinforcement…

Building Reinforcement Learning (RL) algorithms which are able to adapt to continuously evolving tasks is an open research challenge. One technology that is known to inherently handle such non-stationary input patterns well is Hierarchical…

Can a lightweight Vision Transformer (ViT) match or exceed the performance of Convolutional Neural Networks (CNNs) like ResNet on small datasets with small image resolutions? This report demonstrates that a pure ViT can indeed achieve…

In this paper, we address the issue of how to enhance the generalization performance of convolutional neural networks (CNN) in the early learning stage for image classification. This is motivated by real-time applications that require the…

Hyperparameter selection generally relies on running multiple full training trials, with selection based on validation set performance. We propose a gradient-based approach for locally adjusting hyperparameters during training of the model.…

Reversibility in artificial neural networks allows us to retrieve the input given an output. We present feature alignment, a method for approximating reversibility in arbitrary neural networks. We train a network by minimizing the distance…

This paper presents our proposed approach that won the first prize at the ICLR competition on Hardware Aware Efficient Training. The challenge is to achieve the highest possible accuracy in an image classification task in less than 10…

The use of meta-learning and transfer learning in the task of few-shot image classification is a well researched area with many papers showcasing the advantages of transfer learning over meta-learning in cases where data is plentiful and…