Yicong Liu

Yicong Liu, Kaili Wang, Patrick Loa et al.

The COVID-19 pandemic dramatically catalyzed the proliferation of e-shopping. The dramatic growth of e-shopping will undoubtedly cause significant impacts on travel demand. As a result, transportation modeller's ability to model e-shopping demand is becoming increasingly important. This study developed models to predict household' weekly home delivery frequencies. We used both classical econometric and machine learning techniques to obtain the best model. It is found that socioeconomic factors such as having an online grocery membership, household members' average age, the percentage of male household members, the number of workers in the household and various land use factors influence home delivery demand. This study also compared the interpretations and performances of the machine learning models and the classical econometric model. Agreement is found in the variable's effects identified through the machine learning and econometric models. However, with similar recall accuracy, the ordered probit model, a classical econometric model, can accurately predict the aggregate distribution of household delivery demand. In contrast, both machine learning models failed to match the observed distribution.

Yixiong Zou, Yicong Liu, Yiman Hu et al.

Cross-domain few-shot learning (CDFSL) aims to acquire knowledge from limited training data in the target domain by leveraging prior knowledge transferred from source domains with abundant training samples. CDFSL faces challenges in transferring knowledge across dissimilar domains and fine-tuning models with limited training data. To address these challenges, we initially extend the analysis of loss landscapes from the parameter space to the representation space, which allows us to simultaneously interpret the transferring and fine-tuning difficulties of CDFSL models. We observe that sharp minima in the loss landscapes of the representation space result in representations that are hard to transfer and fine-tune. Moreover, existing flatness-based methods have limited generalization ability due to their short-range flatness. To enhance the transferability and facilitate fine-tuning, we introduce a simple yet effective approach to achieve long-range flattening of the minima in the loss landscape. This approach considers representations that are differently normalized as minima in the loss landscape and flattens the high-loss region in the middle by randomly sampling interpolated representations. We implement this method as a new normalization layer that replaces the original one in both CNNs and ViTs. This layer is simple and lightweight, introducing only a minimal number of additional parameters. Experimental results on 8 datasets demonstrate that our approach outperforms state-of-the-art methods in terms of average accuracy. Moreover, our method achieves performance improvements of up to 9\% compared to the current best approaches on individual datasets. Our code will be released.

Yaze Zhao, Yicong Liu, Yixiong Zou et al.

Cross-Domain Few-Shot Learning (CDFSL) aims to adapt large-scale pretrained models to specialized target domains with limited samples, yet the few-shot fine-tuning of vision-language models like CLIP remains underexplored. By establishing multiple fine-tuning baselines of CLIP for CDFSL, we find adapter-based methods (e.g., LoRA) consistently outperform prompt-based ones (e.g., MaPLe), contrary to in-domain scenarios. To make those effective in-domain methods competitive again in CDFSL, we analyze this phenomenon and discover LoRA's superiority stems from rectifying the collapsed attention of visual CLS token, enhancing modality alignment and class separation by focusing on text-related visual regions. Further, we find textual EOS token exhibit much better attention to visual samples, and CLIP's standard contrastive loss weakly constrains modality alignment. Based on these insights, we propose Semantic Probe, a plug-and-play attention rectification framework for both adapter- and prompt-based methods. Extensive experiments on four CDFSL benchmarks validate our rationale, achieving state-of-the-art performance and benefiting both fine-tuning paradigms. Codes will be released.

Zecheng Li, Zhihui Cao, Wenke Huang et al.

Graphical user interface (GUI) agents are rapidly progressing toward autonomous interaction and reliable task execution across diverse applications. However, two central challenges remain unresolved: automating the evaluation of agent trajectories and generating high-quality training data at scale to enable continual improvement. Existing approaches often depend on manual annotation or static rule-based verification, which restricts scalability and limits adaptability in dynamic environments. We present MagicGUI-RMS, a multi-agent reward model system that delivers adaptive trajectory evaluation, corrective feedback, and self-evolving learning capabilities. MagicGUI-RMS integrates a Domain-Specific Reward Model (DS-RM) with a General-Purpose Reward Model (GP-RM), enabling fine-grained action assessment and robust generalization across heterogeneous GUI tasks. To support reward learning at scale, we design a structured data construction pipeline that automatically produces balanced and diverse reward datasets, effectively reducing annotation costs while maintaining sample fidelity. During execution, the reward model system identifies erroneous actions, proposes refined alternatives, and continuously enhances agent behavior through an automated data-reflux mechanism. Extensive experiments demonstrate that MagicGUI-RMS yields substantial gains in task accuracy, behavioral robustness. These results establish MagicGUI-RMS as a principled and effective foundation for building self-improving GUI agents driven by reward-based adaptation.