A Comprehensive Evaluation of Quantized Instruction-Tuned Large Language Models: An Experimental Analysis up to 405B

Prior research works have evaluated quantized LLMs using limited metrics such as perplexity or a few basic knowledge tasks and old datasets. Additionally, recent large-scale models such as Llama 3.1 with up to 405B have not been thoroughly examined. This paper evaluates the performance of instruction-tuned LLMs across various quantization methods (GPTQ, AWQ, SmoothQuant, and FP8) on models ranging from 7B to 405B. Using 13 benchmarks, we assess performance across six task types: commonsense Q\&A, knowledge and language understanding, instruction following, hallucination detection, mathematics, and dialogue. Our key findings reveal that (1) quantizing a larger LLM to a similar size as a smaller FP16 LLM generally performs better across most benchmarks, except for hallucination detection and instruction following; (2) performance varies significantly with different quantization methods, model size, and bit-width, with weight-only methods often yielding better results in larger models; (3) task difficulty does not significantly impact accuracy degradation due to quantization; and (4) the MT-Bench evaluation method has limited discriminatory power among recent high-performing LLMs.

Mixed Non-linear Quantization for Vision Transformers

The majority of quantization methods have been proposed to reduce the model size of Vision Transformers, yet most of them have overlooked the quantization of non-linear operations. Only a few works have addressed quantization for non-linear operations, but they applied a single quantization method across all non-linear operations. We believe that this can be further improved by employing a different quantization method for each non-linear operation. Therefore, to assign the most error-minimizing quantization method from the known methods to each non-linear layer, we propose a mixed non-linear quantization that considers layer-wise quantization sensitivity measured by SQNR difference metric. The results show that our method outperforms I-BERT, FQ-ViT, and I-ViT in both 8-bit and 6-bit settings for ViT, DeiT, and Swin models by an average of 0.6%p and 19.6%p, respectively. Our method outperforms I-BERT and I-ViT by 0.6%p and 20.8%p, respectively, when training time is limited. We plan to release our code at https://gitlab.com/ones-ai/mixed-non-linear-quantization.

Positioning Using Wireless Networks: Applications, Recent Progress and Future Challenges

Positioning has recently received considerable attention as a key enabler in emerging applications such as extended reality, unmanned aerial vehicles and smart environments. These applications require both data communication and high-precision positioning, and thus they are particularly well-suited to be offered in wireless networks (WNs). The purpose of this paper is to provide a comprehensive overview of existing works and new trends in the field of positioning techniques from both the academic and industrial perspectives. The paper provides a comprehensive overview of positioning in WNs, covering the background, applications, measurements, state-of-the-art technologies and future challenges. The paper outlines the applications of positioning from the perspectives of public facilities, enterprises and individual users. We investigate the key performance indicators and measurements of positioning systems, followed by the review of the key enabler techniques such as artificial intelligence/large models and adaptive systems. Next, we discuss a number of typical wireless positioning technologies. We extend our overview beyond the academic progress, to include the standardization efforts, and finally, we provide insight into the challenges that remain. The comprehensive overview of exisitng efforts and new trends in the field of positioning from both the academic and industrial communities would be a useful reference to researchers in the field.

Visual Preference Inference: An Image Sequence-Based Preference Reasoning in Tabletop Object Manipulation

In robotic object manipulation, human preferences can often be influenced by the visual attributes of objects, such as color and shape. These properties play a crucial role in operating a robot to interact with objects and align with human intention. In this paper, we focus on the problem of inferring underlying human preferences from a sequence of raw visual observations in tabletop manipulation environments with a variety of object types, named Visual Preference Inference (VPI). To facilitate visual reasoning in the context of manipulation, we introduce the Chain-of-Visual-Residuals (CoVR) method. CoVR employs a prompting mechanism that describes the difference between the consecutive images (i.e., visual residuals) and incorporates such texts with a sequence of images to infer the user's preference. This approach significantly enhances the ability to understand and adapt to dynamic changes in its visual environment during manipulation tasks. Furthermore, we incorporate such texts along with a sequence of images to infer the user's preferences. Our method outperforms baseline methods in terms of extracting human preferences from visual sequences in both simulation and real-world environments. Code and videos are available at: \href{https://joonhyung-lee.github.io/vpi/}{https://joonhyung-lee.github.io/vpi/}

Q-HyViT: Post-Training Quantization for Hybrid Vision Transformer with Bridge Block Reconstruction

Recently, vision transformers (ViT) have replaced convolutional neural network models in numerous tasks, including classification, detection, and segmentation. However, the high computational requirements of ViTs hinder their widespread implementation. To address this issue, researchers have proposed efficient hybrid transformer architectures that combine convolutional and transformer layers and optimize attention computation for linear complexity. Additionally, post-training quantization has been proposed as a means of mitigating computational demands. Combining quantization techniques and efficient hybrid transformer structures is crucial to maximize the acceleration of vision transformers on mobile devices. However, no prior investigation has applied quantization to efficient hybrid transformers. In this paper, at first, we discover that the straightforward manner to apply the existing PTQ methods for ViT to efficient hybrid transformers results in a drastic accuracy drop due to the following challenges: (i) highly dynamic ranges, (ii) zero-point overflow, (iii) diverse normalization, and (iv) limited model parameters (<5M). To overcome these challenges, we propose a new post-training quantization method, which is the first to quantize efficient hybrid vision transformers (MobileViTv1 and MobileViTv2) with a significant margin (an average improvement of 7.75%) compared to existing PTQ methods (EasyQuant, FQ-ViT, and PTQ4ViT). We plan to release our code at https://github.com/Q-HyViT.

Average Age of Information Penalty of Short-Packet Communications with Packet Management

In this paper, we analyze the non-linear age of information (AoI) performance in a point-to-point short packet communication system, where a transmitter generates packets based on status updates and transmits the packets to a receiver. Specifically, we investigate three packet management strategies, namely, the non-preemption with no buffer strategy, the non-preemption with one buffer strategy, and the preemption strategy. To characterize the level of the receiver's dissatisfaction on outdated data, we adopt a generalized \alpha-\beta AoI penalty function into the analysis and derive closed-form expressions for the average AoI penalty achieved by the three packet management strategies. Simulation results are used to corroborate our analysis and explicitly evaluate the impact of various system parameters, such as the coding rate and status update generation rate, on the AoI performance. Additionally, we find that the value of \alpha reflects the system transmission reliability.

CPrune: Compiler-Informed Model Pruning for Efficient Target-Aware DNN Execution

Mobile devices run deep learning models for various purposes, such as image classification and speech recognition. Due to the resource constraints of mobile devices, researchers have focused on either making a lightweight deep neural network (DNN) model using model pruning or generating an efficient code using compiler optimization. Surprisingly, we found that the straightforward integration between model compression and compiler auto-tuning often does not produce the most efficient model for a target device. We propose CPrune, a compiler-informed model pruning for efficient target-aware DNN execution to support an application with a required target accuracy. CPrune makes a lightweight DNN model through informed pruning based on the structural information of subgraphs built during the compiler tuning process. Our experimental results show that CPrune increases the DNN execution speed up to 2.73x compared to the state-of-the-art TVM auto-tune while satisfying the accuracy requirement.

Quantune: Post-training Quantization of Convolutional Neural Networks using Extreme Gradient Boosting for Fast Deployment

To adopt convolutional neural networks (CNN) for a range of resource-constrained targets, it is necessary to compress the CNN models by performing quantization, whereby precision representation is converted to a lower bit representation. To overcome problems such as sensitivity of the training dataset, high computational requirements, and large time consumption, post-training quantization methods that do not require retraining have been proposed. In addition, to compensate for the accuracy drop without retraining, previous studies on post-training quantization have proposed several complementary methods: calibration, schemes, clipping, granularity, and mixed-precision. To generate a quantized model with minimal error, it is necessary to study all possible combinations of the methods because each of them is complementary and the CNN models have different characteristics. However, an exhaustive or a heuristic search is either too time-consuming or suboptimal. To overcome this challenge, we propose an auto-tuner known as Quantune, which builds a gradient tree boosting model to accelerate the search for the configurations of quantization and reduce the quantization error. We evaluate and compare Quantune with the random, grid, and genetic algorithms. The experimental results show that Quantune reduces the search time for quantization by approximately 36.5x with an accuracy loss of 0.07 ~ 0.65% across six CNN models, including the fragile ones (MobileNet, SqueezeNet, and ShuffleNet). To support multiple targets and adopt continuously evolving quantization works, Quantune is implemented on a full-fledged compiler for deep learning as an open-sourced project.

Energy-efficient Cooperative Offloading for Edge Computing-enabled Vehicular Networks

Edge computing technology has great potential to improve various computation-intensive applications in vehicular networks by providing sufficient computation resources for vehicles. However, it is still a challenge to fully unleash the potential of edge computing in edge computing-enabled vehicular networks. In this paper, we develop the energy-efficient cooperative offloading scheme for edge computing-enabled vehicular networks, which splits the task into multiple subtasks and offloads them to different roadside units (RSUs) located ahead along the route of the vehicle. We first establish novel cooperative offloading models for the offline and online scenarios in edge computing-enabled vehicular networks. In each offloading scenario, we formulate the total energy minimization with respect to the task splitting ratio, computation resource, and communication resource. In the offline scenario, we equivalently transform the original problem to a convex problem and obtain optimal solutions for multi-vehicle case and single-vehicle case, respectively. Furthermore, we show that the method proposed for the offline scenario can also be applied to solve the optimization problem in the online scenario. Finally, through numerical results, by analyzing the impact of network parameters on the total energy consumption, we verify that our proposed solution consumes lower energy than baseline schemes.

Precoding Design for Multi-user MIMO Systems with Delay-Constrained and -Tolerant Users

In both academia and industry, multi-user multiple-input multiple-output (MU-MIMO) techniques have shown enormous gains in spectral efficiency by exploiting spatial degrees of freedom. So far, an underlying assumption in most of the existing MU-MIMO design has been that all the users use infinite blocklength, so that they can achieve the Shannon capacity. This setup, however, is not suitable considering delay-constrained users whose blocklength tends to be finite. In this paper, we consider a heterogeneous setting in MU-MIMO systems where delay-constrained users and delay-tolerant users coexist, called a DCTU-MIMO network. To maximize the sum spectral efficiency in this system, we present the spectral efficiency for delay-tolerant users and provide a lower bound of the spectral efficiency for delay-constrained users. We consider an optimization problem that maximizes the sum spectral efficiency of delay-tolerant users while satisfying the latency constraint of delay-constrained users, and propose a generalized power iteration (GPI) precoding algorithm that finds a principal precoding vector. Furthermore, we extend a DCTU-MIMO network to the multiple time slots scenario and propose a recursive generalized power iteration precoding algorithm. In simulation results, we prove proposed methods outperform baseline schemes and present the effect of network parameters on the ergodic sum spectral efficiency.