In this paper, we demonstrate how to leverage 2:4 sparsity, a popular hardwareaccelerated GPU sparsity pattern, to activations to accelerate large language model training and inference. Crucially we exploit the intrinsic sparsity found in Squared-ReLU activations to provide this acceleration with no accuracy loss. Our approach achieves up to 1.3x faster Feed Forward Network (FFNs) in both the forwards and backwards pass. This work highlights the potential for sparsity to play a key role in accelerating large language model training and inference. The rapid growth of Large Language Models (LLMs) in recent years has been driven by a corresponding surge in GPU FLOPs.

We present Movie Gen, a cast of foundation models that generates high-quality, 1080p HD videos with different aspect ratios and synchronized audio. We also show additional capabilities such as precise instruction-based video editing and generation of personalized videos based on a user's image. Our models set a new state-of-the-art on multiple tasks: text-to-video synthesis, video personalization, video editing, video-to-audio generation, and text-to-audio generation. Our largest video generation model is a 30B parameter transformer trained with a maximum context length of 73K video tokens, corresponding to a generated video of 16 seconds at 16 frames-per-second. We show multiple technical innovations and simplifications on the architecture, latent spaces, training objectives and recipes, data curation, evaluation protocols, parallelization techniques, and inference optimizations that allow us to reap the benefits of scaling pre-training data, model size, and training compute for training large scale media generation models. We hope this paper helps the research community to accelerate progress and innovation in media generation models. All videos from this paper are available at https://go.fb.me/MovieGenResearchVideos.

Post-training quantization (PTQ) techniques applied to weights, activations, and the KV cache greatly reduce memory usage, latency, and power consumption of Large Language Models (LLMs), but may lead to large quantization errors when outliers are present. Recent findings suggest that rotating activation or weight matrices helps remove outliers and benefits quantization. In this work, we identify a collection of applicable rotation parameterizations that lead to identical outputs in full-precision Transformer architectures, and find that some random rotations lead to much better quantization than others, with an up to 13 points difference in downstream zero-shot reasoning performance. As a result, we propose SpinQuant that optimizes (or learns) the rotation matrices with Cayley optimization on a small validation set. With 4-bit quantization of weight, activation, and KV-cache, SpinQuant narrows the accuracy gap on zero-shot reasoning tasks with full precision to merely 2.9 points on the LLaMA-2 7B model, surpassing LLM-QAT by 19.1 points and SmoothQuant by 25.0 points. SpinQuant also outperforms concurrent work QuaRot, which applies random rotations to remove outliers. In particular, for LLaMA-2 7B/LLaMA-3 8B models that are hard to quantize, SpinQuant reduces the gap to full precision by 30.2%/34.1% relative to QuaRot.

Narrow bit-width data formats are key to reducing the computational and storage costs of modern deep learning applications. This paper evaluates Microscaling (MX) data formats that combine a per-block scaling factor with narrow floating-point and integer types for individual elements. MX formats balance the competing needs of hardware efficiency, model accuracy, and user friction. Empirical results on over two dozen benchmarks demonstrate practicality of MX data formats as a drop-in replacement for baseline FP32 for AI inference and training with low user friction. We also show the first instance of training generative language models at sub-8-bit weights, activations, and gradients with minimal accuracy loss and no modifications to the training recipe.

We present RecD (Recommendation Deduplication), a suite of end-to-end infrastructure optimizations across the Deep Learning Recommendation Model (DLRM) training pipeline. RecD addresses immense storage, preprocessing, and training overheads caused by feature duplication inherent in industry-scale DLRM training datasets. Feature duplication arises because DLRM datasets are generated from interactions. While each user session can generate multiple training samples, many features' values do not change across these samples. We demonstrate how RecD exploits this property, end-to-end, across a deployed training pipeline. RecD optimizes data generation pipelines to decrease dataset storage and preprocessing resource demands and to maximize duplication within a training batch. RecD introduces a new tensor format, InverseKeyedJaggedTensors (IKJTs), to deduplicate feature values in each batch. We show how DLRM model architectures can leverage IKJTs to drastically increase training throughput. RecD improves the training and preprocessing throughput and storage efficiency by up to 2.48x, 1.79x, and 3.71x, respectively, in an industry-scale DLRM training system.

Sequence-based deep learning recommendation models (DLRMs) are an emerging class of DLRMs showing great improvements over their prior sum-pooling based counterparts at capturing users' long term interests. These improvements come at immense system cost however, with sequence-based DLRMs requiring substantial amounts of data to be dynamically materialized and communicated by each accelerator during a single iteration. To address this rapidly growing bottleneck, we present FlexShard, a new tiered sequence embedding table sharding algorithm which operates at a per-row granularity by exploiting the insight that not every row is equal. Through precise replication of embedding rows based on their underlying probability distribution, along with the introduction of a new sharding strategy adapted to the heterogeneous, skewed performance of real-world cluster network topologies, FlexShard is able to significantly reduce communication demand while using no additional memory compared to the prior state-of-the-art. When evaluated on production-scale sequence DLRMs, FlexShard was able to reduce overall global all-to-all communication traffic by over 85%, resulting in end-to-end training communication latency improvements of nearly 6x over the prior state-of-the-art approach.

Data Parallelism (DP) and Model Parallelism (MP) are two common paradigms to enable large-scale distributed training of neural networks. Recent trends, such as the improved model performance of deeper and wider neural networks when trained with billions of data points, have prompted the use of hybrid parallelism---a paradigm that employs both DP and MP to scale further parallelization for machine learning. Hybrid training allows compute power to increase, but it runs up against the key bottleneck of communication overhead that hinders scalability. In this paper, we propose a compression framework called Dynamic Communication Thresholding (DCT) for communication-efficient hybrid training. DCT filters the entities to be communicated across the network through a simple hard-thresholding function, allowing only the most relevant information to pass through. For communication efficient DP, DCT compresses the parameter gradients sent to the parameter server during model synchronization, while compensating for the introduced errors with known techniques. For communication efficient MP, DCT incorporates a novel technique to compress the activations and gradients sent across the network during the forward and backward propagation, respectively. This is done by identifying and updating only the most relevant neurons of the neural network for each training sample in the data. Under modest assumptions, we show that the convergence of training is maintained with DCT. We evaluate DCT on natural language processing and recommender system models. DCT reduces overall communication by 20x, improving end-to-end training time on industry scale models by 37%. Moreover, we observe an improvement in the trained model performance, as the induced sparsity is possibly acting as an implicit sparsity based regularization.

Ever growing volume and velocity of data coupled with decreasing attention span of end users underscore the critical need for real-time analytics. In this regard, anomaly detection plays a key role as an application as well as a means to verify data fidelity. Although the subject of anomaly detection has been researched for over 100 years in a multitude of disciplines such as, but not limited to, astronomy, statistics, manufacturing, econometrics, marketing, most of the existing techniques cannot be used as is on real-time data streams. Further, the lack of characterization of performance -- both with respect to real-timeliness and accuracy -- on production data sets makes model selection very challenging. To this end, we present an in-depth analysis, geared towards real-time streaming data, of anomaly detection techniques. Given the requirements with respect to real-timeliness and accuracy, the analysis presented in this paper should serve as a guide for selection of the "best" anomaly detection technique. To the best of our knowledge, this is the first characterization of anomaly detection techniques proposed in very diverse set of fields, using production data sets corresponding to a wide set of application domains.