Author name cluster

Mengwei Li

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

4 papers

1 author row

AAAI Conference 2026 Conference Paper

PosPrune: Visual Token Pruning with Positional Bias Correction for Efficient Large Vision-Language Models

Ziyang Wang
Mengwei Li
Hao Yin
Wenhao Liu
Zilei Wang

Large Vision-Language Models (LVLMs) enhance performance on vision-language tasks by integrating visual features from pre-trained vision encoders into large language models (LLMs). However, the large number of visual tokens introduces significant computational overhead. Existing token pruning methods either perform global selection via [CLS]-based attention in the vision encode or prune within LLM decoding layers. These approaches face two key challenges: (1) [CLS]-based attention primarily focuses on visually salient regions across the entire image, often overlooking semantically important tokens essential for reasoning; and (2) strong positional bias in the shallow decoder layers causes the model to favor later-positioned tokens, while neglecting earlier ones that may carry critical reasoning cues. To address these issues, we propose PosPrune, a training-free, two-stage visual token pruning framework. At the vision encoder, we introduce an Asymmetric Region-aware Pruning (ARP) strategy that retains more tokens in semantically rich regions while discarding more tokens from semantically less informative regions, thus preserving spatial diversity and task-relevant details. In the LLM decoding stage, we find that the positional bias in shallow layers is primarily driven by model architecture rather than task semantics. Based on this insight, we propose a novel Positional Bias Correction (PBC) mechanism to mitigate this bias. To further reduce redundancy, we apply Maximal Marginal Relevance (MMR) to select tokens that best balance textual relevance and diversity. Extensive experiments on various LVLMs and benchmarks demonstrate the general effectiveness of our approach. Notably, when applied to LLaVA-1.5-7B, PosPrune achieves a reduction of 85% in FLOPs while preserving 98.5% of the original performance.

PDF Details DOI

AAAI Conference 2026 Conference Paper

Rethinking Open-world Prompt Tuning: A Systematic Framework for Evaluation and Optimization

Mengwei Li
Zilei Wang
Yixin Zhang

Prompt Tuning (PT) is a widely used strategy for adapting pre-trained Vision-Language Models (VLMs) to various downstream tasks. Conventional PT methods evaluate performance separately on known (base) and unknown (new) classes. However, in real-world scenarios, models often encounter inputs without prior knowledge of their class domain. This challenge has motivated the development of Open-world Prompt Tuning (OPT), which requires models to first determine whether a sample belongs to base or new classes and then classify it accordingly. In this work, we carefully review existing OPT methods and identify three key limitations: (L1) incomplete evaluation metrics, (L2) time-consuming and memory-intensive OOD detection methods, and (L3) insufficiently comprehensive optimization strategies. To address these issues, we first tackle L1 by proposing two novel metrics to explicitly evaluate adaptability and generalization under the OPT setting, forming a more comprehensive evaluation framework. For L2, we propose a training-free OOD detection method called Entropy-weighted Rank-normalized Fusion (ERF), which first applies rank normalization to both the maximum and the sum of base-class probabilities, followed by an entropy-weighted fusion of the normalized values. For L3, we propose a plug-and-play Gated Dual-Merging (GDM) strategy to strengthen the classifier’s capability. GDM performs selective merging at the weight level based on an adaptive criterion and combines fine-tuned and LLM-boosted logits at the output level. Extensive experiments on three PT baselines across 11 datasets demonstrate the effectiveness of our proposed ERF and GDM.

PDF Details DOI

EAAI Journal 2023 Journal Article

DGFaceNet: Lightweight and efficient face recognition

Feng Zhao
Peng Zhang
Ran Zhang
Mengwei Li

Face recognition has achieved great success due to the development of deep convolutional neural networks (DCNNs). However, complex DCNNs bring a large number of parameters as well as computational effort, which poses a significant challenge to resource-constrained embedded devices. Meanwhile, the commonly popular loss functions and lightweight networks are not so effective for face recognition. In this paper, we first investigate the impact of the number of similar features generated by inexpensive operations on model performance. It is shown that DCNNs can tolerate more similar features generated by cheap operations in the early stage of the network. We construct Dynamic Ghost Bottleneck based on this idea, and DGFaceNet is composed of stacking Dynamic Ghost Bottleneck. In addition, we propose a new class-margin-linear softmax loss function (CML-softmax) for lightweight networks. CML-softmax designs a quadratic function to replace the cosine function as the target logit, which allows better performance and convergence in low-dimensional output for face recognition. Meanwhile, CML-softmax introduces two margin functions to alleviate class imbalance and softmax early saturation problems, respectively. Our method demonstrates competitive results in many validation datasets and large-scale popular benchmark tests. Speed tests on embedded devices show that the actual inference time of DGFaceNet is 11. 08 times, 8. 57 times, 2. 75 times, and 2. 82 times faster than ResNet-50, EfficientNet, MobileNetV2, and MobileFaceNet, respectively. DGFaceNet can significantly improve the running efficiency of the model in resource-constrained embedded devices while ensuring the model’s performance.

Details DOI

YNICL Journal 2022 Journal Article

Effects of acute high intraocular pressure on red-green and blue-yellow cortical color responses in non-human primates

Mengwei Li
Xiaoxiao Chen
Nini Yuan
Yiliang Lu
Ye Liu
Hongliang Gong
Liling Qian
Ian Max Andolina

Glaucoma is a leading cause of irreversible blindness worldwide, and intraocular pressure (IOP) is an established and modifiable risk factor for both chronic and acute glaucoma. The relationship between color vision deficits and chronic glaucoma has been described previously. However, the effects of acute glaucoma or acute primary angle closure, which has high prevalence in China, on color vision remains unclear. To address the above question, red-green or blue-yellow color responses in V1, V2, and V4 of seven rhesus macaques were monitored using intrinsic-signal optical imaging while monocular anterior chamber perfusions were performed to reversibly elevate IOP acutely over a clinically observed range of 30 to 90 mmHg. We found that the cortical population responses to both red-green and blue-yellow grating stimuli, systematically decreased as IOP increased from 30 to 90 mmHg. Although a similar decrement in magnitude was noted in V1, V2, and V4, blue-yellow responses were consistently more impaired than red-green responses at all levels of acute IOP elevation and in all monitored visual areas. This physiological study in non-human primates demonstrates that acute IOP elevations substantially depress the ability of the visual cortex to register color information. This effect is more severe for blue-yellow responses than for red-green responses, suggesting selective impairment of the koniocellular pathways compared with the parvocellular pathways. Together, we infer that blue-yellow color vision might be the most vulnerable visual function in acute glaucoma patients.

Details DOI