Wenshuo Li

Jialiang Kang, Han Shu, Wenshuo Li et al.

Speculative decoding is a widely adopted technique for accelerating inference in large language models (LLMs), yet its application to vision-language models (VLMs) remains underexplored, with existing methods achieving only modest speedups (<1.5x). This gap is increasingly significant as multimodal capabilities become central to large-scale models. We hypothesize that large VLMs can effectively filter redundant image information layer by layer without compromising textual comprehension, whereas smaller draft models struggle to do so. To address this, we introduce Vision-Aware Speculative Decoding (ViSpec), a novel framework tailored for VLMs. ViSpec employs a lightweight vision adaptor module to compress image tokens into a compact representation, which is seamlessly integrated into the draft model's attention mechanism while preserving original image positional information. Additionally, we extract a global feature vector for each input image and augment all subsequent text tokens with this feature to enhance multimodal coherence. To overcome the scarcity of multimodal datasets with long assistant responses, we curate a specialized training dataset by repurposing existing datasets and generating extended outputs using the target VLM with modified prompts. Our training strategy mitigates the risk of the draft model exploiting direct access to the target model's hidden states, which could otherwise lead to shortcut learning when training solely on target model outputs. Extensive experiments validate ViSpec, achieving, to our knowledge, the first substantial speedup in VLM speculative decoding. Code is available at https://github.com/KangJialiang/ViSpec.

Xuefei Ning, Junbo Zhao, Wenshuo Li et al.

Convolutional neural networks (CNNs) are vulnerable to adversarial examples, and studies show that increasing the model capacity of an architecture topology (e.g., width expansion) can bring consistent robustness improvements. This reveals a clear robustness-efficiency trade-off that should be considered in architecture design. In this paper, considering scenarios with capacity budget, we aim to discover adversarially robust architecture at targeted capacities. Recent studies employed one-shot neural architecture search (NAS) to discover robust architectures. However, since the capacities of different topologies cannot be aligned in the search process, one-shot NAS methods favor topologies with larger capacities in the supernet. And the discovered topology might be suboptimal when augmented to the targeted capacity. We propose a novel multi-shot NAS method to address this issue and explicitly search for robust architectures at targeted capacities. At the targeted FLOPs of 2000M, the discovered MSRobNet-2000 outperforms the recent NAS-discovered architecture RobNet-large under various criteria by a large margin of 4%-7%. And at the targeted FLOPs of 1560M, MSRobNet-1560 surpasses another NAS-discovered architecture RobNet-free by 2.3% and 1.3% in the clean and PGD-7 accuracies, respectively. All codes are available at https://github.com/walkerning/aw\_nas.

Xuefei Ning, Changcheng Tang, Wenshuo Li et al.

Conducting efficient performance estimations of neural architectures is a major challenge in neural architecture search (NAS). To reduce the architecture training costs in NAS, one-shot estimators (OSEs) amortize the architecture training costs by sharing the parameters of one "supernet" between all architectures. Recently, zero-shot estimators (ZSEs) that involve no training are proposed to further reduce the architecture evaluation cost. Despite the high efficiency of these estimators, the quality of such estimations has not been thoroughly studied. In this paper, we conduct an extensive and organized assessment of OSEs and ZSEs on five NAS benchmarks: NAS-Bench-101/201/301, and NDS ResNet/ResNeXt-A. Specifically, we employ a set of NAS-oriented criteria to study the behavior of OSEs and ZSEs and reveal that they have certain biases and variances. After analyzing how and why the OSE estimations are unsatisfying, we explore how to mitigate the correlation gap of OSEs from several perspectives. Through our analysis, we give out suggestions for future application and development of efficient architecture performance estimators. Furthermore, the analysis framework proposed in our work could be utilized in future research to give a more comprehensive understanding of newly designed architecture performance estimators. All codes are available at https://github.com/walkerning/aw_nas.

Xuefei Ning, Tianchen Zhao, Wenshuo Li et al.

Budgeted pruning is the problem of pruning under resource constraints. In budgeted pruning, how to distribute the resources across layers (i.e., sparsity allocation) is the key problem. Traditional methods solve it by discretely searching for the layer-wise pruning ratios, which lacks efficiency. In this paper, we propose Differentiable Sparsity Allocation (DSA), an efficient end-to-end budgeted pruning flow. Utilizing a novel differentiable pruning process, DSA finds the layer-wise pruning ratios with gradient-based optimization. It allocates sparsity in continuous space, which is more efficient than methods based on discrete evaluation and search. Furthermore, DSA could work in a pruning-from-scratch manner, whereas traditional budgeted pruning methods are applied to pre-trained models. Experimental results on CIFAR-10 and ImageNet show that DSA could achieve superior performance than current iterative budgeted pruning methods, and shorten the time cost of the overall pruning process by at least 1.5x in the meantime.

Huizi Mao, Song Han, Jeff Pool et al.

Sparsity helps reduce the computational complexity of deep neural networks by skipping zeros. Taking advantage of sparsity is listed as a high priority in next generation DNN accelerators such as TPU. The structure of sparsity, i.e., the granularity of pruning, affects the efficiency of hardware accelerator design as well as the prediction accuracy. Coarse-grained pruning creates regular sparsity patterns, making it more amenable for hardware acceleration but more challenging to maintain the same accuracy. In this paper we quantitatively measure the trade-off between sparsity regularity and prediction accuracy, providing insights in how to maintain accuracy while having more a more structured sparsity pattern. Our experimental results show that coarse-grained pruning can achieve a sparsity ratio similar to unstructured pruning without loss of accuracy. Moreover, due to the index saving effect, coarse-grained pruning is able to obtain a better compression ratio than fine-grained sparsity at the same accuracy threshold. Based on the recent sparse convolutional neural network accelerator (SCNN), our experiments further demonstrate that coarse-grained sparsity saves about 2x the memory references compared to fine-grained sparsity. Since memory reference is more than two orders of magnitude more expensive than arithmetic operations, the regularity of sparse structure leads to more efficient hardware design.