Quantization-Aware and Tensor-Compressed Training of Transformers for Natural Language Understanding
Fine-tuned transformer models have shown superior performances in many natural language tasks. However, the large model size prohibits deploying high-performance transformer models on resource-constrained devices. This paper proposes a quantization-aware tensor-compressed training approach to reduce the model size, arithmetic operations, and ultimately runtime latency of transformer-based models. We compress the embedding and linear layers of transformers into small low-rank tensor cores, which significantly reduces model parameters. A quantization-aware training with learnable scale factors is used to further obtain low-precision representations of the tensor-compressed models. The developed approach can be used for both end-to-end training and distillation-based training. To improve the convergence, a layer-by-layer distillation is applied to distill a quantized and tensor-compressed student model from a pre-trained transformer. The performance is demonstrated in two natural language understanding tasks, showing up to $63\times$ compression ratio, little accuracy loss and remarkable inference and training speedup.
相關內容
ACM/IEEE第23屆模型驅動工程語言和系統國際會議，是模型驅動軟件和系統工程的首要會議系列，由ACM-SIGSOFT和IEEE-TCSE支持組織。自1998年以來，模型涵蓋了建模的各個方面，從語言和方法到工具和應用程序。模特的參加者來自不同的背景，包括研究人員、學者、工程師和工業專業人士。MODELS 2019是一個論壇，參與者可以圍繞建模和模型驅動的軟件和系統交流前沿研究成果和創新實踐經驗。今年的版本將為建模社區提供進一步推進建模基礎的機會，并在網絡物理系統、嵌入式系統、社會技術系統、云計算、大數據、機器學習、安全、開源等新興領域提出建模的創新應用以及可持續性。
官網鏈接： ·
樣例
·
數據集
·
標注
·
MoDELS
·
Deep learning models have shown promising predictive accuracy for time-series healthcare applications. However, ensuring the robustness of these models is vital for building trustworthy AI systems. Existing research predominantly focuses on robustness to synthetic adversarial examples, crafted by adding imperceptible perturbations to clean input data. However, these synthetic adversarial examples do not accurately reflect the most challenging real-world scenarios, especially in the context of healthcare data. Consequently, robustness to synthetic adversarial examples may not necessarily translate to robustness against naturally occurring adversarial examples, which is highly desirable for trustworthy AI. We propose a method to curate datasets comprised of natural adversarial examples to evaluate model robustness. The method relies on probabilistic labels obtained from automated weakly-supervised labeling that combines noisy and cheap-to-obtain labeling heuristics. Based on these labels, our method adversarially orders the input data and uses this ordering to construct a sequence of increasingly adversarial datasets. Our evaluation on six medical case studies and three non-medical case studies demonstrates the efficacy and statistical validity of our approach to generating naturally adversarial datasets
Delay alignment modulation (DAM) is a novel wideband transmission technique for mmWave massive MIMO systems, which exploits the high spatial resolution and multi-path sparsity to mitigate ISI, without relying on channel equalization or multi-carrier transmission. In particular, DAM leverages the delay pre-compensation and path-based beamforming to effectively align the multi-path components, thus achieving the constructive multi-path combination for eliminating the ISI while preserving the multi-path power gain. Different from the existing works only considering single-user DAM, this paper investigates the DAM technique for multi-user mmWave massive MIMO communication. First, we consider the asymptotic regime when the number of antennas Mt at BS is sufficiently large. It is shown that by employing the simple delay pre-compensation and per-path-based MRT beamforming, the single-carrier DAM is able to perfectly eliminate both ISI and IUI. Next, we consider the general scenario with Mt being finite. In this scenario, we characterize the achievable rate region of the multi-user DAM system by finding its Pareto boundary. Specifically, we formulate a rate-profile-constrained sum rate maximization problem by optimizing the per-path-based beamforming. Furthermore, we present three low-complexity per-path-based beamforming strategies based on the MRT, zero-forcing, and regularized zero-forcing principles, respectively, based on which the achievable sum rates are studied. Finally, we provide simulation results to demonstrate the performance of our proposed strategies as compared to two benchmark schemes based on the strongest-path-based beamforming and the prevalent OFDM, respectively. It is shown that DAM achieves higher spectral efficiency and/or lower peak-to-average-ratio, for systems with high spatial resolution and multi-path diversity.
We present DictaBERT, a new state-of-the-art pre-trained BERT model for modern Hebrew, outperforming existing models on most benchmarks. Additionally, we release two fine-tuned versions of the model, designed to perform two specific foundational tasks in the analysis of Hebrew texts: prefix segmentation and morphological tagging. These fine-tuned models allow any developer to perform prefix segmentation and morphological tagging of a Hebrew sentence with a single call to a HuggingFace model, without the need to integrate any additional libraries or code. In this paper we describe the details of the training as well and the results on the different benchmarks. We release the models to the community, along with sample code demonstrating their use. We release these models as part of our goal to help further research and development in Hebrew NLP.
We propose TCSP, a novel method for compressing a transformer model by focusing on reducing the hidden size of the model. By projecting the whole transform model into a subspace, we enable matrix operations between the weight matrices in the model and features in a reduced-dimensional space, leading to significant reductions in model parameters and computing resources. To establish this subspace, we decompose the feature matrix, derived from different layers of sampled data instances, into a projection matrix. For evaluation, TCSP is applied to compress T5 and BERT models on the GLUE and SQuAD benchmarks. Experimental results demonstrate that TCSP achieves a compression ratio of 44\% with at most 1.6\% degradation in accuracy, surpassing or matching prior compression methods. Furthermore, TCSP exhibits compatibility with other methods targeting filter and attention head size compression.
Multi-modal Large Language Model (MLLM) refers to a model expanded from a Large Language Model (LLM) that possesses the capability to handle and infer multi-modal data. Current MLLMs typically begin by using LLMs to decompose tasks into multiple subtasks, then employing individual pre-trained models to complete specific subtasks, and ultimately utilizing LLMs to integrate the results of each subtasks to obtain the results of the task. In real-world scenarios, when dealing with large projects, it is common practice to break down the project into smaller sub-projects, with different teams providing corresponding solutions or results. The project owner then decides which solution or result to use, ensuring the best possible outcome for each subtask and, consequently, for the entire project. Inspired by this, this study considers selecting multiple pre-trained models to complete the same subtask. By combining the results from multiple pre-trained models, the optimal subtask result is obtained, enhancing the performance of the MLLM. Specifically, this study first selects multiple pre-trained models focused on the same subtask based on distinct evaluation approaches, and then invokes these models in parallel to process input data and generate corresponding subtask results. Finally, the results from multiple pre-trained models for the same subtask are compared using the LLM, and the best result is chosen as the outcome for that subtask. Extensive experiments are conducted in this study using GPT-4 annotated datasets and human-annotated datasets. The results of various evaluation metrics adequately demonstrate the effectiveness of the proposed approach in this paper.
Pre-trained language models like ChatGPT have significantly improved code generation. As these models scale up, there is an increasing need for the output to handle more intricate tasks. Moreover, in bioinformatics, generating functional programs poses additional notable challenges due to the amount of domain knowledge, the need for complicated data operations, and intricate functional dependencies between the operations. Here, we present BioCoder, a benchmark developed to evaluate existing pre-trained models in generating bioinformatics code. In relation to function-code generation, BioCoder covers potential package dependencies, class declarations, and global variables. It incorporates 1026 functions and 1243 methods in Python and Java from GitHub and 253 examples from the Rosalind Project. BioCoder incorporates a fuzz-testing framework for evaluation, and we have applied it to evaluate many models including InCoder, CodeGen, CodeGen2, SantaCoder, StarCoder, StarCoder+, InstructCodeT5+, and ChatGPT. Our detailed analysis of these models emphasizes the importance of domain knowledge, pragmatic code generation, and contextual understanding. Our dataset, benchmark, Docker images, and scripts required for testing are all available at //github.com/gersteinlab/biocoder.
Generative models are now capable of producing highly realistic images that look nearly indistinguishable from the data on which they are trained. This raises the question: if we have good enough generative models, do we still need datasets? We investigate this question in the setting of learning general-purpose visual representations from a black-box generative model rather than directly from data. Given an off-the-shelf image generator without any access to its training data, we train representations from the samples output by this generator. We compare several representation learning methods that can be applied to this setting, using the latent space of the generator to generate multiple "views" of the same semantic content. We show that for contrastive methods, this multiview data can naturally be used to identify positive pairs (nearby in latent space) and negative pairs (far apart in latent space). We find that the resulting representations rival those learned directly from real data, but that good performance requires care in the sampling strategy applied and the training method. Generative models can be viewed as a compressed and organized copy of a dataset, and we envision a future where more and more "model zoos" proliferate while datasets become increasingly unwieldy, missing, or private. This paper suggests several techniques for dealing with visual representation learning in such a future. Code is released on our project page: //ali-design.github.io/GenRep/
Deep neural networks (DNNs) are successful in many computer vision tasks. However, the most accurate DNNs require millions of parameters and operations, making them energy, computation and memory intensive. This impedes the deployment of large DNNs in low-power devices with limited compute resources. Recent research improves DNN models by reducing the memory requirement, energy consumption, and number of operations without significantly decreasing the accuracy. This paper surveys the progress of low-power deep learning and computer vision, specifically in regards to inference, and discusses the methods for compacting and accelerating DNN models. The techniques can be divided into four major categories: (1) parameter quantization and pruning, (2) compressed convolutional filters and matrix factorization, (3) network architecture search, and (4) knowledge distillation. We analyze the accuracy, advantages, disadvantages, and potential solutions to the problems with the techniques in each category. We also discuss new evaluation metrics as a guideline for future research.
Automatic KB completion for commonsense knowledge graphs (e.g., ATOMIC and ConceptNet) poses unique challenges compared to the much studied conventional knowledge bases (e.g., Freebase). Commonsense knowledge graphs use free-form text to represent nodes, resulting in orders of magnitude more nodes compared to conventional KBs (18x more nodes in ATOMIC compared to Freebase (FB15K-237)). Importantly, this implies significantly sparser graph structures - a major challenge for existing KB completion methods that assume densely connected graphs over a relatively smaller set of nodes. In this paper, we present novel KB completion models that can address these challenges by exploiting the structural and semantic context of nodes. Specifically, we investigate two key ideas: (1) learning from local graph structure, using graph convolutional networks and automatic graph densification and (2) transfer learning from pre-trained language models to knowledge graphs for enhanced contextual representation of knowledge. We describe our method to incorporate information from both these sources in a joint model and provide the first empirical results for KB completion on ATOMIC and evaluation with ranking metrics on ConceptNet. Our results demonstrate the effectiveness of language model representations in boosting link prediction performance and the advantages of learning from local graph structure (+1.5 points in MRR for ConceptNet) when training on subgraphs for computational efficiency. Further analysis on model predictions shines light on the types of commonsense knowledge that language models capture well.
Many tasks in natural language processing can be viewed as multi-label classification problems. However, most of the existing models are trained with the standard cross-entropy loss function and use a fixed prediction policy (e.g., a threshold of 0.5) for all the labels, which completely ignores the complexity and dependencies among different labels. In this paper, we propose a meta-learning method to capture these complex label dependencies. More specifically, our method utilizes a meta-learner to jointly learn the training policies and prediction policies for different labels. The training policies are then used to train the classifier with the cross-entropy loss function, and the prediction policies are further implemented for prediction. Experimental results on fine-grained entity typing and text classification demonstrate that our proposed method can obtain more accurate multi-label classification results.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191