Related papers: CAMERO: Consistency Regularized Ensemble of Pertur…
Subword regularizations use multiple subword segmentations during training to improve the robustness of neural machine translation models. In previous subword regularizations, we use multiple segmentations in the training process but use…
Ensemble methods for supervised machine learning have become popular due to their ability to accurately predict class labels with groups of simple, lightweight "base learners." While ensembles offer computationally efficient models that…
While most current approaches rely on further training techniques, such as fine-tuning or reinforcement learning, to enhance model capacities, model merging stands out for its ability of improving models without requiring any additional…
Model merging aims to integrate the strengths of multiple fine-tuned models into a unified model while preserving task-specific capabilities. Existing methods, represented by task arithmetic, are typically classified into global- and…
Diffusion probabilistic models have achieved mainstream success in many generative modeling tasks, from image generation to inverse problem solving. A distinct feature of these models is that they correspond to deep hierarchical latent…
Model merging aggregates Large Language Models (LLMs) finetuned on different tasks into a stronger one. However, parameter conflicts between models leads to performance degradation in averaging. While model routing addresses this issue by…
Model merging provides a way of cheaply combining individual models to produce a model that inherits each individual's capabilities. While some merging methods can approach the performance of multitask training, they are often heuristically…
Machine learning-based Deepfake detection models have achieved impressive results on benchmark datasets, yet their performance often deteriorates significantly when evaluated on out-of-distribution data. In this work, we investigate an…
We investigate the effectiveness of ensembles of pretrained transformer-based language models on short answer questions using the Kaggle Automated Short Answer Scoring dataset. We fine-tune a collection of popular small, base, and large…
Transformer-based pre-trained language models (PLMs) mostly suffer from excessive overhead despite their advanced capacity. For resource-constrained devices, there is an urgent need for a spatially and temporally efficient model which…
Deep learning-based video compression is a challenging task, and many previous state-of-the-art learning-based video codecs use optical flows to exploit the temporal correlation between successive frames and then compress the residual…
Memory-augmented neural networks consisting of a neural controller and an external memory have shown potentials in long-term sequential learning. Current RAM-like memory models maintain memory accessing every timesteps, thus they do not…
In this work, we explore whether modeling recurrence into the Transformer architecture can both be beneficial and efficient, by building an extremely simple recurrent module into the Transformer. We compare our model to baselines following…
Neural language models do not scale well when the vocabulary is large. Noise-contrastive estimation (NCE) is a sampling-based method that allows for fast learning with large vocabularies. Although NCE has shown promising performance in…
Language models can learn a range of capabilities from unsupervised training on text corpora. However, to solve a particular problem (such as text summarization) it is typically necessary to fine-tune them on a task-specific dataset. It is…
Many deep learning applications benefit from using large models with billions of parameters. Training these models is notoriously expensive due to the need for specialized HPC clusters. In this work, we consider alternative setups for…
Group imbalance has been a known problem in empirical risk minimization (ERM), where the achieved high average accuracy is accompanied by low accuracy in a minority group. Despite algorithmic efforts to improve the minority group accuracy,…
Quantization has significantly improved the compute and memory efficiency of Large Language Model (LLM) training. However, existing approaches still rely on accumulating their updates in high-precision: concretely, gradient updates must be…
In this work, we study the problem of continual learning (CL) where the goal is to learn a model on a sequence of tasks, such that the data from the previous tasks becomes unavailable while learning on the current task data. CL is…
We propose a simple and efficient approach for training the BERT model. Our approach exploits the special structure of BERT that contains a stack of repeated modules (i.e., transformer encoders). Our proposed approach first trains BERT with…