Yinghuan Shi

Is Better Teacher Supervision Enough? Unlocking Student-side Learning in Multimodal On-Policy Distillation

Siyuan Liu, Kanghui Tian, Yue Duan +4

On-policy distillation (OPD) improves reasoning by providing token-level supervision from a teacher on a student's own trajectories. Existing methods primarily focus on enhancing t…

cs.LG2026

Exploring Dualistic Meta-Learning to Enhance Domain Generalization in Open Set Scenarios

Xiran Wang, Jian Zhang, Lei Qi +2

Domain generalization learns from multiple source domains to generalize to unseen target domains. However, it often neglects the realistic case of label mismatch between source and…

cs.CL2026

Are Tools Always Beneficial? Learning to Invoke Tools Adaptively for Dual-Mode Multimodal LLM Reasoning

Qinghe Ma, Zhen Zhao, Yiming Wu +3

Tool-augmented reasoning has emerged as a promising direction for enhancing the reasoning capabilities of multimodal large language models (MLLMs). However, existing studies mainly…

StableMind: Source-Free Cross-Subject fMRI Decoding with Regularized Adaptation

Jintao Guo, Lin Wang, Shumeng Li +5

Existing cross-subject fMRI decoding methods typically train a model on multiple scanned subjects and then adapt it to a new subject using substantial paired fMRI-image data. Howev…

One Token, Two Fates: A Unified Framework via Vision Token Manipulation Against MLLMs Hallucination

Zhan Fa, Yue Duan, Jian Zhang +2

Current training-free methods tackle MLLM hallucination with separate strategies: either enhancing visual signals or suppressing text inertia. However, these separate methods are i…

Duala: Dual-Level Alignment of Subjects and Stimuli for Cross-Subject fMRI Decoding

Shumeng Li, Jintao Guo, Jian Zhang +3

Cross-subject visual decoding aims to reconstruct visual experiences from brain activity across individuals, enabling more scalable and practical brain-computer interfaces. However…

cs.LG2026

When Shared Knowledge Hurts: Spectral Over-Accumulation in Model Merging

Yayuan Li, Ze Peng, Jian Zhang +3

Model merging combines multiple fine-tuned models into a single model by adding their weight updates, providing a lightweight alternative to retraining. Existing methods primarily…

cs.LG2026

Decomposing and Composing: Towards Efficient Vision-Language Continual Learning via Rank-1 Expert Pool in a Single LoRA

Zhan Fa, Yue Duan, Jian Zhang +3

Continual learning (CL) in vision-language models (VLMs) faces significant challenges in improving task adaptation and avoiding catastrophic forgetting. Existing methods usually ha…

MAGIC: Achieving Superior Model Merging via Magnitude Calibration

Yayuan Li, Jian Zhang, Jintao Guo +4

The proliferation of pre-trained models has given rise to a wide array of specialised, fine-tuned models. Model merging aims to merge the distinct capabilities of these specialised…

On the Implicit Adversariality of Catastrophic Forgetting in Deep Continual Learning

Ze Peng, Jian Zhang, Jintao Guo +3

Continual learning seeks the human-like ability to accumulate new skills in machine intelligence. Its central challenge is catastrophic forgetting, whose underlying cause has not b…

Diversity-enhanced Collaborative Mamba for Semi-supervised Medical Image Segmentation

Shumeng Li, Jian Zhang, Lei Qi +3

Acquiring high-quality annotated data for medical image segmentation is tedious and costly. Semi-supervised segmentation techniques alleviate this burden by leveraging unlabeled da…

Towards Perfection: Building Inter-component Mutual Correction for Retinex-based Low-light Image Enhancement

Luyang Cao, Han Xu, Jian Zhang +4

In low-light image enhancement, Retinex-based deep learning methods have garnered significant attention due to their exceptional interpretability. These methods decompose images in…

Correspondence as Video: Test-Time Adaption on SAM2 for Reference Segmentation in the Wild

Haoran Wang, Zekun Li, Jian Zhang +2

Large vision models like the Segment Anything Model (SAM) exhibit significant limitations when applied to downstream tasks in the wild. Consequently, reference segmentation, which…

Divide-and-Conquer for Enhancing Unlabeled Learning, Stability, and Plasticity in Semi-supervised Continual Learning

Yue Duan, Taicai Chen, Lei Qi +1

Semi-supervised continual learning (SSCL) seeks to leverage both labeled and unlabeled data in a sequential learning setup, aiming to reduce annotation costs while managing continu…

Unleashing the Power of Intermediate Domains for Mixed Domain Semi-Supervised Medical Image Segmentation

Qinghe Ma, Jian Zhang, Lei Qi +3

Both limited annotation and domain shift are prevalent challenges in medical image segmentation. Traditional semi-supervised segmentation and unsupervised domain adaptation methods…

Background Matters: A Cross-view Bidirectional Modeling Framework for Semi-supervised Medical Image Segmentation

Luyang Cao, Jianwei Li, Yinghuan Shi

Semi-supervised medical image segmentation (SSMIS) leverages unlabeled data to reduce reliance on manually annotated images. However, current SOTA approaches predominantly focus on…

Balanced Direction from Multifarious Choices: Arithmetic Meta-Learning for Domain Generalization

Xiran Wang, Jian Zhang, Lei Qi +1

Domain generalization is proposed to address distribution shift, arising from statistical disparities between training source and unseen target domains. The widely used first-order…

Taste More, Taste Better: Diverse Data and Strong Model Boost Semi-Supervised Crowd Counting

Maochen Yang, Zekun Li, Jian Zhang +2

Semi-supervised crowd counting is crucial for addressing the high annotation costs of densely populated scenes. Although several methods based on pseudo-labeling have been proposed…

Steady Progress Beats Stagnation: Mutual Aid of Foundation and Conventional Models in Mixed Domain Semi-Supervised Medical Image Segmentation

Qinghe Ma, Jian Zhang, Zekun Li +3

Large pretrained visual foundation models exhibit impressive general capabilities. However, the extensive prior knowledge inherent in these models can sometimes be a double-edged s…