Recent advancements in large language models (LLMs) have extended their capabilities to handle long contexts. However, increasing the number of model layers and the length of input sequences significantly escalates the memory required to store key-value (KV) cache, posing challenges for efficient inference. To mitigate this issue, we present SimLayerKV, a simple yet effective method that reduces inter-layer KV cache redundancies by selectively dropping cache in identified lazy layers. Our approach is based on the observation that certain layers in long-context LLMs exhibit "lazy" behavior, contributing less to modeling long-range dependencies compared to non-lazy layers. By analyzing attention weight patterns, we find that the behavior of these lazy layers is consistent across tokens during generation for a given input. This insight motivates our SimLayerKV, which identifies lazy layers and reduces their KV cache accordingly. SimLayerKV is training-free, generalizable, and can be implemented with only seven lines of code. We conduct extensive experiments on three representative LLMs, e.g., LLaMA2-7B, LLaMA3-8B, and Mistral-7B across 16 tasks from the LongBench benchmark. The results demonstrate that SimLayerKV achieves a KV cache compression ratio of 5$\times$ with only a 1.2% performance drop when combined with 4-bit quantization. Our code is available at https://github.com/sail-sg/SimLayerKV.
Transformers serve as the backbone architectures of Foundational Models, where domain-specific tokenizers allow them to adapt to various domains. Graph Transformers (GTs) have recently emerged as leading models in geometric deep learning, outperforming Graph Neural Networks (GNNs) in various graph learning tasks. However, the development of tokenizers for graphs has lagged behind other modalities. To address this, we introduce GQT (\textbf{G}raph \textbf{Q}uantized \textbf{T}okenizer), which decouples tokenizer training from Transformer training by leveraging multi-task graph self-supervised learning, yielding robust and generalizable graph tokens. Furthermore, the GQT utilizes Residual Vector Quantization (RVQ) to learn hierarchical discrete tokens, resulting in significantly reduced memory requirements and improved generalization capabilities. By combining the GQT with token modulation, a Transformer encoder achieves state-of-the-art performance on 20 out of 22 benchmarks, including large-scale homophilic and heterophilic datasets.
Prompt Tuning has been a popular Parameter-Efficient Fine-Tuning method attributed to its remarkable performance with few updated parameters on various large-scale pretrained Language Models (PLMs). Traditionally, each prompt has been considered indivisible and updated independently, leading the parameters increase proportionally as prompt length grows. To address this issue, we propose Adaptive Codebook for Composite and Efficient Prompt Tuning (ACCEPT). In our method, we refer to the concept of product quantization (PQ), allowing all soft prompts to share a set of learnable codebook vectors in each subspace, with each prompt differentiated by a set of adaptive weights. We achieve the superior performance on 17 diverse natural language tasks including natural language understanding (NLU) and question answering (QA) tasks by tuning only 0.3% of parameters of the PLMs. Our approach also excels in few-shot and large model settings, highlighting its significant potential.
Fairness is an important objective throughout society. From the distribution of limited goods such as education, over hiring and payment, to taxes, legislation, and jurisprudence. Due to the increasing importance of machine learning approaches in all areas of daily life including those related to health, security, and equity, an increasing amount of research focuses on fair machine learning. In this work, we focus on the fairness of partition- and prototype-based models. The contribution of this work is twofold: 1) we develop a general framework for fair machine learning of partition-based models that does not depend on a specific fairness definition, and 2) we derive a fair version of learning vector quantization (LVQ) as a specific instantiation. We compare the resulting algorithm against other algorithms from the literature on theoretical and real-world data showing its practical relevance.
Large language models (LLMs) excel in various tasks but face deployment challenges due to hardware constraints. We propose density-aware post-training weight-only quantization (DAQ), which has two stages: 1) density-centric alignment, which identifies the center of high-density weights and centers the dynamic range on this point to align high-density weight regions with floating-point high-precision regions; 2) learnable dynamic range adjustment, which adjusts the dynamic range by optimizing quantization parameters (i.e., scale and zero-point) based on the impact of weights on the model output. Experiments on LLaMA and LLaMA-2 show that DAQ consistently outperforms the best baseline method, reducing perplexity loss by an average of 22.8% on LLaMA and 19.6% on LLaMA-2. Our code is available at https://github.com/LuoYingSong/DAQ.
Quantization reduces the model's hardware costs, such as data movement, storage, and operations like multiply and addition. It also affects the model's behavior by degrading the output quality. Therefore, there is a need for methods that preserve the model's behavior when quantizing model parameters. More exotic numerical encodings, such as block-scaled number formats, have shown advantages for utilizing a fixed bit budget to encode model parameters. This paper presents error diffusion (ED), a hyperparameter-free method for post-training quantization with support for block-scaled data formats. Our approach does not rely on backpropagation or Hessian information. We describe how to improve the quantization process by viewing the neural model as a composite function and diffusing the quantization error in every layer. In addition, we introduce TensorCast, an open-source library based on PyTorch to emulate a variety of number formats, including the block-scaled ones, to aid the research in neural model quantization. We demonstrate the efficacy of our algorithm through rigorous testing on various architectures, including vision and large language models (LLMs), where it consistently delivers competitive results. Our experiments confirm that block-scaled data formats provide a robust choice for post-training quantization and could be used effectively to enhance the practical deployment of advanced neural networks.
The edge-device environment imposes severe resource limitations, encompassing computation costs, hardware resource usage, and energy consumption for deploying deep neural network models. Ultra-low-bit quantization and hardware accelerators have been explored as promising approaches to address these challenges. Ultra-low-bit quantization significantly reduces the model size and the computational cost. Despite progress so far, many competitive ultra-low-bit models still partially rely on float or non-ultra-low-bit quantized computation such as the input and output layer. We introduce Efficiera Residual Networks (ERNs), a model optimized for low-resource edge devices. ERNs achieve full ultra-low-bit quantization, with all weights, including the initial and output layers, being binary, and activations set at 2 bits. We introduce the shared constant scaling factor technique to enable integer-valued computation in residual connections, allowing our model to operate without float values until the final convolution layer. Demonstrating competitiveness, ERNs achieve an ImageNet top-1 accuracy of 72.5pt with a ResNet50-compatible architecture and 63.6pt with a model size less than 1MB. Moreover, ERNs exhibit impressive inference times, reaching 300FPS with the smallest model and 60FPS with the largest model on a cost-efficient FPGA device.
Deploying long-context large language models (LLMs) is essential but poses significant computational and memory challenges. Caching all Key and Value (KV) states across all attention heads consumes substantial memory. Existing KV cache pruning methods either damage the long-context capabilities of LLMs or offer only limited efficiency improvements. In this paper, we identify that only a fraction of attention heads, a.k.a, Retrieval Heads, are critical for processing long contexts and require full attention across all tokens. In contrast, all other heads, which primarily focus on recent tokens and attention sinks--referred to as Streaming Heads--do not require full attention. Based on this insight, we introduce DuoAttention, a framework that only applies a full KV cache to retrieval heads while using a light-weight, constant-length KV cache for streaming heads, which reduces both LLM's decoding and pre-filling memory and latency without compromising its long-context abilities. DuoAttention uses a lightweight, optimization-based algorithm with synthetic data to identify retrieval heads accurately. Our method significantly reduces long-context inference memory by up to 2.55x for MHA and 1.67x for GQA models while speeding up decoding by up to 2.18x and 1.50x and accelerating pre-filling by up to 1.73x and 1.63x for MHA and GQA models, respectively, with minimal accuracy loss compared to full attention. Notably, combined with quantization, DuoAttention enables Llama-3-8B decoding with 3.3 million context length on a single A100 GPU. Code is provided in https://github.com/mit-han-lab/duo-attention.
Recently, quantization has been widely used for the compression and acceleration of large language models~(LLMs). Due to the outliers in LLMs, it is crucial to flatten weights and activations to minimize quantization error with the equally spaced quantization points. Prior research explores various pre-quantization transformations to suppress outliers, such as per-channel scaling and Hadamard transformation. However, we observe that these transformed weights and activations can still remain steep and outspread. In this paper, we propose FlatQuant (Fast and Learnable Affine Transformation), a new post-training quantization approach to enhance flatness of weights and activations. Our approach identifies optimal affine transformations tailored to each linear layer, calibrated in hours via a lightweight objective. To reduce runtime overhead, we apply Kronecker decomposition to the transformation matrices, and fuse all operations in FlatQuant into a single kernel. Extensive experiments show that FlatQuant sets up a new state-of-the-art quantization benchmark. For instance, it achieves less than $\textbf{1}\%$ accuracy drop for W4A4 quantization on the LLaMA-3-70B model, surpassing SpinQuant by $\textbf{7.5}\%$. For inference latency, FlatQuant reduces the slowdown induced by pre-quantization transformation from 0.26x of QuaRot to merely $\textbf{0.07x}$, bringing up to $\textbf{2.3x}$ speedup for prefill and $\textbf{1.7x}$ speedup for decoding, respectively. Code is available at: \url{https://github.com/ruikangliu/FlatQuant}.
Conventional model compression techniques for LLMs address high memory consumption and slow inference challenges but typically require computationally expensive retraining to preserve accuracy. In contrast, one-shot compression methods eliminate retraining cost, but struggle to achieve accuracy comparable to dense models. This paper presents SLIM, a new one-shot compression framework that holistically integrates hardware-friendly quantization, sparsity, and low-rank approximation into a unified process. First, we formulate the quantization process using a probabilistic approach (SLIM-Quant) that enables us to apply uniform quantization. Then, we use an existing one-shot pruning method to apply semi-structured sparsity on top of the quantized weights. Finally, to compensate for the introduced aggregated quantization and sparsity error, we use a novel saliency function with unique invertible and additive features that enables us to mathematically compute the value of low-rank adapters. SLIM improves model accuracy by up to 5.66% (LLaMA-2-7B) for 2:4 sparsity with 4-bit weight quantization, outperforming prior methods. Models compressed with SLIM achieve up to 4.3x and 3.8x on Nvidia RTX3060 and A100 GPUs, respectively. Additionally, they achieve up to 0.23x end-to-end memory reduction in comparison to their dense counterparts. We also propose an optional PEFT recipe that further improves accuracy by up to 1.66% (LLaMA-2-13B) compared to SLIM without fine-tuning.
Language Models (LMs) assign significant attention to the first token, even if it is not semantically important, which is known as attention sink. This phenomenon has been widely adopted in applications such as streaming/long context generation, KV cache optimization, inference acceleration, model quantization, and others. Despite its widespread use, a deep understanding of attention sink in LMs is still lacking. In this work, we first demonstrate that attention sinks exist universally in LMs with various inputs, even in small models. Furthermore, attention sink is observed to emerge during the LM pre-training, motivating us to investigate how optimization, data distribution, loss function, and model architecture in LM pre-training influence its emergence. We highlight that attention sink emerges after effective optimization on sufficient training data. The sink position is highly correlated with the loss function and data distribution. Most importantly, we find that attention sink acts more like key biases, storing extra attention scores, which could be non-informative and not contribute to the value computation. We also observe that this phenomenon (at least partially) stems from tokens' inner dependence on attention scores as a result of softmax normalization. After relaxing such dependence by replacing softmax attention with other attention operations, such as sigmoid attention without normalization, attention sinks do not emerge in LMs up to 1B parameters. The code is available at https://github.com/sail-sg/Attention-Sink.
With the rapid growth in the use of fine-tuning for large language models (LLMs), optimizing fine-tuning while keeping inference efficient has become highly important. However, this is a challenging task as it requires improvements in all aspects, including inference speed, fine-tuning speed, memory consumption, and, most importantly, model quality. Previous studies have attempted to achieve this by combining quantization with fine-tuning, but they have failed to enhance all four aspects simultaneously. In this study, we propose a new lightweight technique called Quantization for Efficient Fine-Tuning (QEFT). QEFT accelerates both inference and fine-tuning, is supported by robust theoretical foundations, offers high flexibility, and maintains good hardware compatibility. Our extensive experiments demonstrate that QEFT matches the quality and versatility of full-precision parameter-efficient fine-tuning, while using fewer resources. Our code is available at https://github.com/xvyaward/qeft.
The spatio-temporal complexity of video data presents significant challenges in tasks such as compression, generation, and inpainting. We present four key contributions to address the challenges of spatiotemporal video processing. First, we introduce the 3D Mobile Inverted Vector-Quantization Variational Autoencoder (3D-MBQ-VAE), which combines Variational Autoencoders (VAEs) with masked token modeling to enhance spatiotemporal video compression. The model achieves superior temporal consistency and state-of-the-art (SOTA) reconstruction quality by employing a novel training strategy with full frame masking. Second, we present MotionAura, a text-to-video generation framework that utilizes vector-quantized diffusion models to discretize the latent space and capture complex motion dynamics, producing temporally coherent videos aligned with text prompts. Third, we propose a spectral transformer-based denoising network that processes video data in the frequency domain using the Fourier Transform. This method effectively captures global context and long-range dependencies for high-quality video generation and denoising. Lastly, we introduce a downstream task of Sketch Guided Video Inpainting. This task leverages Low-Rank Adaptation (LoRA) for parameter-efficient fine-tuning. Our models achieve SOTA performance on a range of benchmarks. Our work offers robust frameworks for spatiotemporal modeling and user-driven video content manipulation. We will release the code, datasets, and models in open-source.
In this paper, we propose a post-training quantization framework of large vision-language models (LVLMs) for efficient multi-modal inference. Conventional quantization methods sequentially search the layer-wise rounding functions by minimizing activation discretization errors, which fails to acquire optimal quantization strategy without considering cross-layer dependency. On the contrary, we mine the cross-layer dependency that significantly influences discretization errors of the entire vision-language model, and embed this dependency into optimal quantization strategy searching with low search cost. Specifically, we observe the strong correlation between the activation entropy and the cross-layer dependency concerning output discretization errors. Therefore, we employ the entropy as the proxy to partition blocks optimally, which aims to achieve satisfying trade-offs between discretization errors and the search cost. Moreover, we optimize the visual encoder to disentangle the cross-layer dependency for fine-grained decomposition of search space, so that the search cost is further reduced without harming the quantization accuracy. Experimental results demonstrate that our method compresses the memory by 2.78x and increase generate speed by 1.44x about 13B LLaVA model without performance degradation on diverse multi-modal reasoning tasks. Code is available at https://github.com/ChangyuanWang17/QVLM.
Recent advances in latent diffusion-based generative models for portrait image animation, such as Hallo, have achieved impressive results in short-duration video synthesis. In this paper, we present updates to Hallo, introducing several design enhancements to extend its capabilities. First, we extend the method to produce long-duration videos. To address substantial challenges such as appearance drift and temporal artifacts, we investigate augmentation strategies within the image space of conditional motion frames. Specifically, we introduce a patch-drop technique augmented with Gaussian noise to enhance visual consistency and temporal coherence over long duration. Second, we achieve 4K resolution portrait video generation. To accomplish this, we implement vector quantization of latent codes and apply temporal alignment techniques to maintain coherence across the temporal dimension. By integrating a high-quality decoder, we realize visual synthesis at 4K resolution. Third, we incorporate adjustable semantic textual labels for portrait expressions as conditional inputs. This extends beyond traditional audio cues to improve controllability and increase the diversity of the generated content. To the best of our knowledge, Hallo2, proposed in this paper, is the first method to achieve 4K resolution and generate hour-long, audio-driven portrait image animations enhanced with textual prompts. We have conducted extensive experiments to evaluate our method on publicly available datasets, including HDTF, CelebV, and our introduced "Wild" dataset. The experimental results demonstrate that our approach achieves state-of-the-art performance in long-duration portrait video animation, successfully generating rich and controllable content at 4K resolution for duration extending up to tens of minutes. Project page https://fudan-generative-vision.github.io/hallo2
Low-resolution analog-to-digital converters (ADCs) have emerged as an efficient solution for massive multiple-input multiple-output (MIMO) systems to reap high data rates with reasonable power consumption and hardware complexity. In this paper, we study precoding designs for digital, fully connected (FC) hybrid, and partially connected (PC) hybrid beamforming architectures in massive MIMO systems with low-resolution ADCs at the receiver. We aim to maximize the spectral efficiency (SE) subject to a transmit power budget and hardware constraints on the analog components. The resulting problems are nonconvex and the quantization distortion introduces additional challenges. To address them, we first derive a tight lower bound for the SE, based on which we optimize the precoders for the three beamforming architectures under the majorization-minorization framework. Numerical results validate the superiority of the proposed precoding designs over their state-of-the-art counterparts in systems with low-resolution ADCs, particularly those with 1-bit resolution. The results show that the PC hybrid precoding design can achieve an SE close to those of the digital and FC hybrid precoding designs in 1-bit systems, highlighting the potential of the PC hybrid beamforming architectures.
Recently, generative semantic communication models have proliferated as they are revolutionizing semantic communication frameworks, improving their performance, and opening the way to novel applications. Despite their impressive ability to regenerate content from the compressed semantic information received, generative models pose crucial challenges for communication systems in terms of high memory footprints and heavy computational load. In this paper, we present a novel Quantized GEnerative Semantic COmmunication framework, Q-GESCO. The core method of Q-GESCO is a quantized semantic diffusion model capable of regenerating transmitted images from the received semantic maps while simultaneously reducing computational load and memory footprint thanks to the proposed post-training quantization technique. Q-GESCO is robust to different channel noises and obtains comparable performance to the full precision counterpart in different scenarios saving up to 75% memory and 79% floating point operations. This allows resource-constrained devices to exploit the generative capabilities of Q-GESCO, widening the range of applications and systems for generative semantic communication frameworks. The code is available at https://github.com/ispamm/Q-GESCO.
Mixture-of-Experts large language models (MoE-LLMs) marks a significant step forward of language models, however, they encounter two critical challenges in practice: 1) expert parameters lead to considerable memory consumption and loading latency; and 2) the current activated experts are redundant, as many tokens may only require a single expert. Motivated by these issues, we investigate the MoE-LLMs and make two key observations: a) different experts exhibit varying behaviors on activation reconstruction error, routing scores, and activated frequencies, highlighting their differing importance, and b) not all tokens are equally important -- only a small subset is critical. Building on these insights, we propose MC-MoE, a training-free Mixture-Compressor for MoE-LLMs, which leverages the significance of both experts and tokens to achieve an extreme compression. First, to mitigate storage and loading overheads, we introduce Pre-Loading Mixed-Precision Quantization, which formulates the adaptive bit-width allocation as a Linear Programming problem, where the objective function balances multi-factors reflecting the importance of each expert. Additionally, we develop Online Dynamic Pruning, which identifies important tokens to retain and dynamically select activated experts for other tokens during inference to optimize efficiency while maintaining performance. Our MC-MoE integrates static quantization and dynamic pruning to collaboratively achieve extreme compression for MoE-LLMs with less accuracy loss, ensuring an optimal trade-off between performance and efficiency. Extensive experiments confirm the effectiveness of our approach. For instance, at 2.54 bits, MC-MoE compresses 76.6% of the model, with only a 3.8% average accuracy loss. During dynamic inference, we further reduce activated parameters by 15%, with a performance drop of less than 0.6%.
Error correction codes (ECC) are crucial for ensuring reliable information transmission in communication systems. Choukroun & Wolf (2022b) recently introduced the Error Correction Code Transformer (ECCT), which has demonstrated promising performance across various transmission channels and families of codes. However, its high computational and memory demands limit its practical applications compared to traditional decoding algorithms. Achieving effective quantization of the ECCT presents significant challenges due to its inherently small architecture, since existing, very low-precision quantization techniques often lead to performance degradation in compact neural networks. In this paper, we introduce a novel acceleration method for transformer-based decoders. We first propose a ternary weight quantization method specifically designed for the ECCT, inducing a decoder with multiplication-free linear layers. We present an optimized self-attention mechanism to reduce computational complexity via codeaware multi-heads processing. Finally, we provide positional encoding via the Tanner graph eigendecomposition, enabling a richer representation of the graph connectivity. The approach not only matches or surpasses ECCT's performance but also significantly reduces energy consumption, memory footprint, and computational complexity. Our method brings transformer-based error correction closer to practical implementation in resource-constrained environments, achieving a 90% compression ratio and reducing arithmetic operation energy consumption by at least 224 times on modern hardware.
Image processing is one of the most promising applications for quantum machine learning (QML). Quanvolutional Neural Networks with non-trainable parameters are the preferred solution to run on current and near future quantum devices. The typical input preprocessing pipeline for quanvolutional layers comprises of four steps: optional input binary quantization, encoding classical data into quantum states, processing the data to obtain the final quantum states, decoding quantum states back to classical outputs. In this paper we propose two ways to enhance the efficiency of quanvolutional models. First, we propose a flexible data quantization approach with memoization, applicable to any encoding method. This allows us to increase the number of quantization levels to retain more information or lower them to reduce the amount of circuit executions. Second, we introduce a new integrated encoding strategy, which combines the encoding and processing steps in a single circuit. This method allows great flexibility on several architectural parameters (e.g., number of qubits, filter size, and circuit depth) making them adjustable to quantum hardware requirements. We compare our proposed integrated model with a classical convolutional neural network and the well-known rotational encoding method, on two different classification tasks. The results demonstrate that our proposed model encoding exhibits a comparable or superior performance to the other models while requiring fewer quantum resources.