arXiv subjects

Yanzhi Wang

Publications and source records attributed to Yanzhi Wang.

At least 55 records · Page 3Linked to original sources

arXiv2026

Toward Adaptive Large Language Models Structured Pruning via Hybrid-grained Weight Importance Assessment

Structured pruning for large language models (LLMs) has garnered significant academic interest due to its ability to efficiently compress and accelerate LLMs by eliminating redundant weight groups at a coarse-grained granularity. Current structured pruning methods for LLMs typically depend on a singular granularity for assessing weight importance, resulting in notable performance degradation in downstream tasks. Intriguingly, our empirical investigations reveal that utilizing unstructured pruning, which achieves better performance retention by pruning weights at a finer granularity, \emph{i.e.}, individual weights, yields significantly varied sparse LLM structures when juxtaposed to structured pruning. This suggests that evaluating both holistic and individual assessment for weight importance is essential for LLM pruning. Building on this insight, we introduce the Hybrid-grained Weight Importance Assessment (HyWIA), a novel method that merges fine-grained and coarse-grained evaluations of weight importance for the pruning of LLMs. Leveraging an attention mechanism, HyWIA adaptively determines the optimal blend of granularity in weight importance assessments in an end-to-end pruning manner. Extensive experiments on LLaMA-V1/V2, Vicuna, Baichuan, and Bloom across various benchmarks demonstrate the effectiveness of HyWIA in pruning LLMs. For example, HyWIA surpasses the cutting-edge LLM-Pruner by an average margin of 2.82% in accuracy across seven downstream tasks when pruning LLaMA-7B by 50%. Code:https://github.com/azuryl/LLM-HWIA

cs.CL↗

arXiv2026

S2DiT: Sandwich Diffusion Transformer for Mobile Streaming Video Generation

Diffusion Transformers (DiTs) have recently improved video generation quality. However, their heavy computational cost makes real-time or on-device generation infeasible. In this work, we introduce S2DiT, a Streaming Sandwich Diffusion Transformer designed for efficient, high-fidelity, and streaming video generation on mobile hardware. S2DiT generates more tokens but maintains efficiency with novel efficient attentions: a mixture of LinConv Hybrid Attention (LCHA) and Stride Self-Attention (SSA). Based on this, we uncover the sandwich design via a budget-aware dynamic programming search, achieving superior quality and efficiency. We further propose a 2-in-1 distillation framework that transfers the capacity of large teacher models (e.g., Wan 2.2-14B) to the compact few-step sandwich model. Together, S2DiT achieves quality on par with state-of-the-art server video models, while streaming at over 10 FPS on an iPhone.

cs.CV↗

arXiv2026

HIERAMP: Coarse-to-Fine Autoregressive Amplification for Generative Dataset Distillation

Dataset distillation often prioritizes global semantic proximity when creating small surrogate datasets for original large-scale ones. However, object semantics are inherently hierarchical. For example, the position and appearance of a bird's eyes are constrained by the outline of its head. Global proximity alone fails to capture how object-relevant structures at different levels support recognition. In this work, we investigate the contributions of hierarchical semantics to effective distilled data. We leverage the vision autoregressive (VAR) model whose coarse-to-fine generation mirrors this hierarchy and propose HIERAMP to amplify semantics at different levels. At each VAR scale, we inject class tokens that dynamically identify salient regions and use their induced maps to guide amplification at that scale. This adds only marginal inference cost while steering synthesis toward discriminative parts and structures. Empirically, we find that semantic amplification leads to more diverse token choices in constructing coarse-scale object layouts. Conversely, at fine scales, the amplification concentrates token usage, increasing focus on object-related details. Across popular dataset distillation benchmarks, HIERAMP consistently improves validation performance without explicitly optimizing global proximity, demonstrating the importance of semantic amplification for effective dataset distillation.

cs.CV↗

arXiv2026

CircuitSense: A Hierarchical MLLM Benchmark Bridging Visual Comprehension and Symbolic Reasoning in Engineering Design Process

Engineering design operates through hierarchical abstraction from system specifications to component implementations, requiring visual understanding coupled with mathematical reasoning at each level. While Multi-modal Large Language Models (MLLMs) excel at natural image tasks, their ability to extract mathematical models from technical diagrams remains unexplored. We present \textbf{CircuitSense}, a comprehensive benchmark evaluating circuit understanding across this hierarchy through 8,006+ problems spanning component-level schematics to system-level block diagrams. Our benchmark uniquely examines the complete engineering workflow: Perception, Analysis, and Design, with a particular emphasis on the critical but underexplored capability of deriving symbolic equations from visual inputs. We introduce a hierarchical synthetic generation pipeline consisting of a grid-based schematic generator and a block diagram generator with auto-derived symbolic equation labels. Comprehensive evaluation of six state-of-the-art MLLMs, including both closed-source and open-source models, reveals fundamental limitations in visual-to-mathematical reasoning. Closed-source models achieve over 85\% accuracy on perception tasks involving component recognition and topology identification, yet their performance on symbolic derivation and analytical reasoning falls below 19\%, exposing a critical gap between visual parsing and symbolic reasoning. Models with stronger symbolic reasoning capabilities consistently achieve higher design task accuracy, confirming the fundamental role of mathematical understanding in circuit synthesis and establishing symbolic reasoning as the key metric for engineering competence.

cs.CV↗

arXiv2026

Effective MoE-based LLM Compression by Exploiting Heterogeneous Inter-Group Experts Routing Frequency and Information Density

Mixture-of-Experts (MoE) based Large Language Models (LLMs) have achieved superior performance, yet the massive memory overhead caused by storing multiple expert networks severely hinders their practical deployment. Singular Value Decomposition (SVD)-based compression has emerged as a promising post-training technique; however, most existing methods apply uniform rank allocation or rely solely on static weight properties. This overlooks the substantial heterogeneity in expert utilization observed in MoE models, where frequent routing patterns and intrinsic information density vary significantly across experts. In this work, we propose RFID-MoE, an effective framework for MoE compression by exploiting heterogeneous Routing Frequency and Information Density. We first introduce a fused metric that combines expert activation frequency with effective rank to measure expert importance, adaptively allocating higher ranks to critical expert groups under a fixed budget. Moreover, instead of discarding compression residuals, we reconstruct them via a parameter-efficient sparse projection mechanism to recover lost information with minimal parameter overhead. Extensive experiments on representative MoE LLMs (e.g., Qwen3, DeepSeekMoE) across multiple compression ratios demonstrate that RFID-MoE consistently outperforms state-of-the-art methods like MoBE and D2-MoE. Notably, RFID-MoE achieves a perplexity of 16.92 on PTB with the Qwen3-30B model at a 60% compression ratio, reducing perplexity by over 8.0 compared to baselines, and improves zero-shot accuracy on HellaSwag by approximately 8%.

cs.LG↗

arXiv2026

Open-Source Multimodal Moxin Models with Moxin-VLM and Moxin-VLA

Recently, Large Language Models (LLMs) have undergone a significant transformation, marked by a rapid rise in both their popularity and capabilities. Leading this evolution are proprietary LLMs like GPT-4 and GPT-o1, which have captured widespread attention in the AI community due to their remarkable performance and versatility. Simultaneously, open-source LLMs, such as LLaMA and Mistral, have made great contributions to the ever-increasing popularity of LLMs due to the ease to customize and deploy the models across diverse applications. Moxin 7B is introduced as a fully open-source LLM developed in accordance with the Model Openness Framework, which moves beyond the simple sharing of model weights to embrace complete transparency in training, datasets, and implementation detail, thus fostering a more inclusive and collaborative research environment that can sustain a healthy open-source ecosystem. To further equip Moxin with various capabilities in different tasks, we develop three variants based on Moxin, including Moxin-VLM, Moxin-VLA, and Moxin-Chinese, which target the vision-language, vision-language-action, and Chinese capabilities, respectively. Experiments show that our models achieve superior performance in various evaluations. We adopt open-source framework and open data for the training. We release our models, along with the available data and code to derive these models.

cs.CL↗

arXiv2026

Cross-Platform Scaling of Vision-Language-Action Models from Edge to Cloud GPUs

Vision-Language-Action (VLA) models have emerged as powerful generalist policies for robotic control, yet their performance scaling across model architectures and hardware platforms, as well as their associated power budgets, remain poorly understood. This work presents an evaluation of five representative VLA models -- spanning state-of-the-art baselines and two newly proposed architectures -- targeting edge and datacenter GPU platforms. Using the LIBERO benchmark, we measure accuracy alongside system-level metrics, including latency, throughput, and peak memory usage, under varying edge power constraints and high-performance datacenter GPU configurations. Our results identify distinct scaling trends: (1) architectural choices, such as action tokenization and model backbone size, strongly influence throughput and memory footprint; (2) power-constrained edge devices exhibit non-linear performance degradation, with some configurations matching or exceeding older datacenter GPUs; and (3) high-throughput variants can be achieved without significant accuracy loss. These findings provide actionable insights when selecting and optimizing VLAs across a range of deployment constraints. Our work challenges current assumptions about the superiority of datacenter hardware for robotic inference.

cs.AI↗

arXiv2026

Q-realign: Piggybacking Realignment on Quantization for Safe and Efficient LLM Deployment

Public large language models (LLMs) are typically safety-aligned during pretraining, yet task-specific fine-tuning required for deployment often erodes this alignment and introduces safety risks. Existing defenses either embed safety recovery into fine-tuning or rely on fine-tuning-derived priors for post-hoc correction, leaving safety recovery tightly coupled with training and incurring high computational overhead and a complex workflow. To address these challenges, we propose \texttt{Q-realign}, a post-hoc defense method based on post-training quantization, guided by an analysis of representational structure. By reframing quantization as a dual-objective procedure for compression and safety, \texttt{Q-realign} decouples safety alignment from fine-tuning and naturally piggybacks into modern deployment pipelines. Experiments across multiple models and datasets demonstrate that our method substantially reduces unsafe behaviors while preserving task performance, with significant reductions in memory usage and GPU hours. Notably, our approach can recover the safety alignment of a fine-tuned 7B LLM on a single RTX 4090 within 40 minutes. Overall, our work provides a practical, turnkey solution for safety-aware deployment.

cs.LG↗

arXiv2026

Advancing time series completion via RFAMoE and MDFF

Recent studies show that using diffusion models for time series signal reconstruction holds great promise. However, such approaches remain largely unexplored in the domain of medical time series. The unique characteristics of the physiological time series signals, such as multivariate, high temporal variability, highly noisy, and artifact-prone, make deep learning-based approaches still challenging for tasks such as imputation. Hence, we propose a novel Mixture of Experts (MoE)-based noise estimator within a score-based diffusion framework. Specifically, the Receptive Field Adaptive MoE (RFAMoE) module is designed to enable each channel to adaptively select desired receptive fields throughout the diffusion process. Moreover, recent literature has found that when generating a physiological signal, performing multiple inferences and averaging the reconstructed signals can effectively reduce reconstruction errors, but at the cost of significant computational and latency overhead. We design a Fusion MoE module and innovatively leverage the nature of MoE module to generate K noise signals in parallel, fuse them using a routing mechanism, and complete signal reconstruction in a single inference step. This design not only improves performance over previous methods but also eliminates the substantial computational cost and latency associated with multiple inference processes. Extensive results demonstrate that our proposed framework consistently outperforms diffusion-based SOTA works on different tasks and datasets.

cs.LG↗

arXiv2025

HDCompression: Hybrid-Diffusion Image Compression for Ultra-Low Bitrates

Image compression under ultra-low bitrates remains challenging for both conventional learned image compression (LIC) and generative vector-quantized (VQ) modeling. Conventional LIC suffers from severe artifacts due to heavy quantization, while generative VQ modeling gives poor fidelity due to the mismatch between learned generative priors and specific inputs. In this work, we propose Hybrid-Diffusion Image Compression (HDCompression), a dual-stream framework that utilizes both generative VQ-modeling and diffusion models, as well as conventional LIC, to achieve both high fidelity and high perceptual quality. Different from previous hybrid methods that directly use pre-trained LIC models to generate low-quality fidelity-preserving information from heavily quantized latent, we use diffusion models to extract high-quality complementary fidelity information from the ground-truth input, which can enhance the system performance in several aspects: improving index map prediction, enhancing the fidelity-preserving output of the LIC stream, and refining conditioned image reconstruction with VQ-latent correction. In addition, our diffusion model is based on a dense representative vector (DRV), which is lightweight with very simple sampling schedulers. Extensive experiments demonstrate that our HDCompression outperforms the previous conventional LIC, generative VQ-modeling, and hybrid frameworks in both quantitative metrics and qualitative visualization, providing balanced robust compression performance at ultra-low bitrates.

cs.CV↗

arXiv2025

ZeroSim: Zero-Shot Analog Circuit Evaluation with Unified Transformer Embeddings

Although recent advancements in learning-based analog circuit design automation have tackled tasks such as topology generation, device sizing, and layout synthesis, efficient performance evaluation remains a major bottleneck. Traditional SPICE simulations are time-consuming, while existing machine learning methods often require topology-specific retraining or manual substructure segmentation for fine-tuning, hindering scalability and adaptability. In this work, we propose ZeroSim, a transformer-based performance modeling framework designed to achieve robust in-distribution generalization across trained topologies under novel parameter configurations and zero-shot generalization to unseen topologies without any fine-tuning. We apply three key enabling strategies: (1) a diverse training corpus of 3.6 million instances covering over 60 amplifier topologies, (2) unified topology embeddings leveraging global-aware tokens and hierarchical attention to robustly generalize to novel circuits, and (3) a topology-conditioned parameter mapping approach that maintains consistent structural representations independent of parameter variations. Our experimental results demonstrate that ZeroSim significantly outperforms baseline models such as multilayer perceptrons, graph neural networks and transformers, delivering accurate zero-shot predictions across different amplifier topologies. Additionally, when integrated into a reinforcement learning-based parameter optimization pipeline, ZeroSim achieves a remarkable speedup (13x) compared to conventional SPICE simulations, underscoring its practical value for a wide range of analog circuit design automation tasks.

cs.LG↗

arXiv2025

Harmony in Divergence: Towards Fast, Accurate, and Memory-efficient Zeroth-order LLM Fine-tuning

Large language models (LLMs) excel across various tasks, but standard first-order (FO) fine-tuning demands considerable memory, significantly limiting real-world deployment. Recently, zeroth-order (ZO) optimization stood out as a promising memory-efficient training paradigm, avoiding backward passes and relying solely on forward passes for gradient estimation, making it attractive for resource-constrained scenarios. However, ZO method lags far behind FO method in both convergence speed and accuracy. To bridge the gap, we introduce a novel layer-wise divergence analysis that uncovers the distinct update pattern of FO and ZO optimization. Aiming to resemble the learning capacity of FO method from the findings, we propose Divergence-driven Zeroth-Order (DiZO) optimization. DiZO conducts divergence-driven layer adaptation by incorporating projections to ZO updates, generating diverse-magnitude updates precisely scaled to layer-wise individual optimization needs. Our results demonstrate that DiZO significantly reduces the needed iterations for convergence without sacrificing throughput, cutting training GPU hours by up to 48\% on various datasets. Moreover, DiZO consistently outperforms the representative ZO baselines in fine-tuning RoBERTa-large, OPT-series, and Llama-series on downstream tasks and, in some cases, even surpasses memory-intensive FO fine-tuning. Our code is released at https://github.com/Skilteee/DiZO.

cs.LG↗

arXiv2025

Domain Adaptation-based Edge Computing for Cross-Conditions Fault Diagnosis

Fault diagnosis of mechanical equipment provides robust support for industrial production. It is worth noting that, the operation of mechanical equipment is accompanied by changes in factors such as speed and load, leading to significant differences in data distribution, which pose challenges for fault diagnosis. Additionally, in terms of application deployment, commonly used cloud-based fault diagnosis methods often encounter issues such as time delays and data security concerns, while common fault diagnosis methods cannot be directly applied to edge computing devices. Therefore, conducting fault diagnosis under cross-operating conditions based on edge computing holds significant research value. This paper proposes a domain-adaptation-based lightweight fault diagnosis framework tailored for edge computing scenarios. Incorporating the local maximum mean discrepancy into knowledge transfer aligns the feature distributions of different domains in a high-dimensional feature space, to discover a common feature space across domains. The acquired fault diagnosis expertise from the cloud-based deep neural network model is transferred to the lightweight edge-based model (edge model) using adaptation knowledge transfer methods. It aims to achieve accurate fault diagnosis under cross-working conditions while ensuring real-time diagnosis capabilities. We utilized the NVIDIA Jetson Xavier NX kit as the edge computing platform and conducted validation experiments on two devices. In terms of diagnostic performance, the proposed method significantly improved diagnostic accuracy, with average increases of 34.44% and 17.33% compared to existing methods, respectively.

cs.DC↗

arXiv2025

LightCache: Memory-Efficient, Training-Free Acceleration for Video Generation

Training-free acceleration has emerged as an advanced research area in video generation based on diffusion models. The redundancy of latents in diffusion model inference provides a natural entry point for acceleration. In this paper, we decompose the inference process into the encoding, denoising, and decoding stages, and observe that cache-based acceleration methods often lead to substantial memory surges in the latter two stages. To address this problem, we analyze the characteristics of inference across different stages and propose stage-specific strategies for reducing memory consumption: 1) Asynchronous Cache Swapping. 2) Feature chunk. 3) Slicing latents to decode. At the same time, we ensure that the time overhead introduced by these three strategies remains lower than the acceleration gains themselves. Compared with the baseline, our approach achieves faster inference speed and lower memory usage, while maintaining quality degradation within an acceptable range. The Code is available at https://github.com/NKUShaw/LightCache .

cs.CV↗

arXiv2025

TSLA: A Task-Specific Learning Adaptation for Semantic Segmentation on Autonomous Vehicles Platform

Autonomous driving platforms encounter diverse driving scenarios, each with varying hardware resources and precision requirements. Given the computational limitations of embedded devices, it is crucial to consider computing costs when deploying on target platforms like the NVIDIA\textsuperscript{\textregistered} DRIVE PX 2. Our objective is to customize the semantic segmentation network according to the computing power and specific scenarios of autonomous driving hardware. We implement dynamic adaptability through a three-tier control mechanism -- width multiplier, classifier depth, and classifier kernel -- allowing fine-grained control over model components based on hardware constraints and task requirements. This adaptability facilitates broad model scaling, targeted refinement of the final layers, and scenario-specific optimization of kernel sizes, leading to improved resource allocation and performance. Additionally, we leverage Bayesian Optimization with surrogate modeling to efficiently explore hyperparameter spaces under tight computational budgets. Our approach addresses scenario-specific and task-specific requirements through automatic parameter search, accommodating the unique computational complexity and accuracy needs of autonomous driving. It scales its Multiply-Accumulate Operations (MACs) for Task-Specific Learning Adaptation (TSLA), resulting in alternative configurations tailored to diverse self-driving tasks. These TSLA customizations maximize computational capacity and model accuracy, optimizing hardware utilization.

cs.CV↗

arXiv2025

H3AE: High Compression, High Speed, and High Quality AutoEncoder for Video Diffusion Models

Autoencoder (AE) is the key to the success of latent diffusion models for image and video generation, reducing the denoising resolution and improving efficiency. However, the power of AE has long been underexplored in terms of network design, compression ratio, and training strategy. In this work, we systematically examine the architecture design choices and optimize the computation distribution to obtain a series of efficient and high-compression video AEs that can decode in real time even on mobile devices. We also propose an omni-training objective to unify the design of plain Autoencoder and image-conditioned I2V VAE, achieving multifunctionality in a single VAE network but with enhanced quality. In addition, we propose a novel latent consistency loss that provides stable improvements in reconstruction quality. Latent consistency loss outperforms prior auxiliary losses including LPIPS, GAN and DWT in terms of both quality improvements and simplicity. H3AE achieves ultra-high compression ratios and real-time decoding speed on GPU and mobile, and outperforms prior arts in terms of reconstruction metrics by a large margin. We finally validate our AE by training a DiT on its latent space and demonstrate fast, high-quality text-to-video generation capability.

cs.CV↗

arXiv2025

Taming Diffusion Transformer for Efficient Mobile Video Generation in Seconds

Diffusion Transformers (DiT) have shown strong performance in video generation tasks, but their high computational cost makes them impractical for resource-constrained devices like smartphones, and practical on-device generation is even more challenging. In this work, we propose a series of novel optimizations to significantly accelerate video generation and enable practical deployment on mobile platforms. First, we employ a highly compressed variational autoencoder (VAE) to reduce the dimensionality of the input data without sacrificing visual quality. Second, we introduce a KD-guided, sensitivity-aware tri-level pruning strategy to shrink the model size to suit mobile platforms while preserving critical performance characteristics. Third, we develop an adversarial step distillation technique tailored for DiT, which allows us to reduce the number of inference steps to four. Combined, these optimizations enable our model to achieve approximately 15 frames per second (FPS) generation speed on an iPhone 16 Pro Max, demonstrating the feasibility of efficient, high-quality video generation on mobile devices.

cs.CV↗

arXiv2025

Collaborative Compression for Large-Scale MoE Deployment on Edge

The Mixture of Experts (MoE) architecture is an important method for scaling Large Language Models (LLMs). It increases model capacity while keeping computation cost low. However, the ultra-large MoE models still have hundreds of billions of parameters, requiring massive memory/storage and leading to difficulties for deployment on resource-constrained edge platforms. Pruning or quantization alone can hardly address the issue, because of the super-aggressive compression ratio with significantly degraded accuracy and output quality. To facilitate the deployment of ultra-large MoEs on edge platforms, we propose a collaborative compression framework by combining expert pruning, mixed-precision quantization, and activation optimization. It can effectively reduce the storage footprint of the ultra-large MoE DeepSeek-V3 from 1.3TB to 103GB, while preserving high output quality with better accuracy than traditional uniform low-bit quantization methods. To the best of our knowledge, we are the first to deploy a compressed model from the ultra-large DeepSeek-V3 on the platform with a strict 128GB total memory limit. Our comprehensive experiments on multiple benchmarks under various memory constraints demonstrate the effectiveness of our method with smaller model sizes and higher accuracy than uniform low-bit quantization methods.

cs.AI↗