Related papers: Virtual Parameter Sharpening: Dynamic Low-Rank Per…

Variational Autoencoders (VAEs) are powerful generative models for learning latent representations. Standard VAEs generate dispersed and unstructured latent spaces by utilizing all dimensions, which limits their interpretability, especially…

Vision-language process reward models (VL-PRMs) are increasingly used to score intermediate reasoning steps and rerank candidates under test-time scaling. However, they often function as black-box judges: a low step score may reflect a…

While Convolutional Neural Networks (CNNs) excel at learning complex latent-space representations, their over-parameterization can lead to overfitting and reduced performance, particularly with limited data. This, alongside their high…

Self-supervised speech representation learning (speech SSL) has demonstrated the benefit of scale in learning rich representations for Automatic Speech Recognition (ASR) with limited paired data, such as wav2vec 2.0. We investigate the…

Current virtual reality systems are typically limited by performance/cost, usability (size), or a combination of both. By using a networked client/server environment, we have solved these limitations for the client. However, in doing so we…

Despite the excellent performance of vision-language pre-trained models (VLPs) on conventional VQA task, they still suffer from two problems: First, VLPs tend to rely on language biases in datasets and fail to generalize to…

Training and fine-tuning large language models (LLMs) come with challenges related to memory and computational requirements due to the increasing size of the model weights and the optimizer states. Various techniques have been developed to…

Sparse variational Gaussian process (SVGP) methods are a common choice for non-conjugate Gaussian process inference because of their computational benefits. In this paper, we improve their computational efficiency by using a dual…

Making deep convolutional neural networks more accurate typically comes at the cost of increased computational and memory resources. In this paper, we reduce this cost by exploiting the fact that the importance of features computed by…

Existing studies in weakly-supervised semantic segmentation (WSSS) using image-level weak supervision have several limitations: sparse object coverage, inaccurate object boundaries, and co-occurring pixels from non-target objects. To…

Most of the current speech data augmentation methods operate on either the raw waveform or the amplitude spectrum of speech. In this paper, we propose a novel speech data augmentation method called PhasePerturbation that operates…

Recent advances in text-based large language models (LLMs), particularly in the GPT series and the o1 model, have demonstrated the effectiveness of scaling both training-time and inference-time compute. However, current state-of-the-art TTS…

Reinforcement Learning (RL) has emerged as the key driver for post-training complex reasoning in Large Language Models (LLMs), yet online RL introduces significant instability and computational overhead. Offline RL offers a compelling…

Several applications in the scientific simulation of physical systems can be formulated as control/optimization problems. The computational models for such systems generally contain hyperparameters, which control solution fidelity and…

We present on-line policy gradient algorithms for computing the locally optimal policy of a constrained, average cost, finite state Markov Decision Process. The stochastic approximation algorithms require estimation of the gradient of the…

The growth of large language models underscores the need for parameter-efficient fine-tuning. Despite its popularity, LoRA encounters storage and computational challenges when deploying multiple task- or user-specific modules. To address…

Low-rank approximation methods such as singular value decomposition (SVD) and its variants (e.g., Fisher-weighted SVD, Activation SVD) have recently emerged as effective tools for neural network compression. In this setting, decomposition…

Reinforcement learning (RL) finetuning has become a key technique for enhancing the reasoning abilities of large language models (LLMs). However, its effectiveness critically depends on the selection of training data. Recent advances…

Applying a pre-trained large model to downstream tasks is prohibitive under resource-constrained conditions. Recent dominant approaches for addressing efficiency issues involve adding a few learnable parameters to the fixed backbone model.…

For 6-DOF (degrees of freedom) interactive virtual acoustic environments (VAEs), the spatial rendering of diffuse late reverberation in addition to early (specular) reflections is important. In the interest of computational efficiency, the…