Xin Tong, Weidong Zhang, Jiaang Li, Haibin Chen, Shilei Liu, Langming Liu, Kangtao Lv, Yujin Yuan, Wenbo Su, Bo Zheng. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Large vision Transformers (ViTs) have achieved competitive performance in various computer vision tasks based on large-scale pre-training. However, large ViTs still remain vulnerable to adversarial examples, emphasizing the necessity of enhancing their adversarial robustness. While adversarial training is an effective defense for deep convolutional models, it often faces scalability issues with large ViTs due to high computational costs. Recent approaches propose robust fine-tuning methods, such as adversarial tuning of low-rank adaptation (LoRA) in ViT, however, they still struggle to match the accuracy of full parameter adversarial fine-tuning. An effective synergy of various defense mechanisms offers a promising approach to enhancing the robustness of ViT, yet this paradigm remains largely underexplored. To address this, we propose hyper adversarial tuning (HyperAT), a meta learning approach, which captures shared defensive knowledge among different methods to improve model robustness efficiently and effectively simultaneously. Specifically, adversarial tuning of each defense method is formulated as a learning task, and a HyperNetwork generates LoRA specific to this defense. Then, a random sampling and tuning strategy is proposed to extract and facilitate the defensive knowledge transfer between different defenses. Finally, diverse LoRAs are merged adaptively to further enhance the adversarial robustness. Experiments on various datasets and model architectures demonstrate that HyperAT significantly enhances the adversarial robustness of pretrained large vision models without excessive computational overhead, establishing a new state-of-the-art benchmark.

Large language models (LLMs) have attracted significant attention due to their impressive general capabilities across diverse downstream tasks.However, without domain-specific optimization, they often underperform on specialized knowledge benchmarks and even produce hallucination.Recent studies show that strategically infusing domain knowledge during pretraining can substantially improve downstream performance.A critical challenge lies in balancing this infusion trade-off: injecting too little domain-specific data yields insufficient specialization, whereas excessive infusion triggers catastrophic forgetting of previously acquired knowledge.In this work, we focus on the phenomenon of memory collapse induced by overinfusion.Through systematic experiments, we make two key observations, i.e. 1) Critical collapse point: each model exhibits a threshold beyond which its knowledge retention capabilities sharply degrade.2) Scale correlation: these collapse points scale consistently with the model's size.Building on these insights, we propose a knowledge infusion scaling law that predicts the optimal amount of domain knowledge to inject into large LLMs by analyzing their smaller counterparts.Extensive experiments across different model sizes and pertaining token budgets validate both the effectiveness and generalizability of our scaling law.

With the increasing use of Large Language Models (LLMs) in fields such as e-commerce, domain-specific concept evaluation benchmarks are crucial for assessing their domain capabilities. Existing LLMs may generate factually incorrect information within the complex e-commerce applications. Therefore, it is necessary to build an e-commerce concept benchmark. Existing benchmarks encounter two primary challenges: (1) handle the heterogeneous and diverse nature of tasks(2) distinguish between generality and specificity within the e-commerce field. To address these problems, we propose ChineseEcomQA, a scalable question-answering benchmark focused on fundamental e-commerce concepts. ChineseEcomQA is built on three core characteristics: Focus on Fundamental Concept, E-commerce Generality and E-commerce Expertise. Fundamental concepts are designed to be applicable across a diverse array of e-commerce tasks, thus addressing the challenge of heterogeneity and diversity. Additionally, by carefully balancing generality and specificity, ChineseEcomQA effectively differentiates between broad e-commerce concepts, allowing for precise validation of domain capabilities. We achieve this through a scalable benchmark construction process that combines LLM validation, Retrieval-Augmented Generation (RAG) validation, and rigorous manual annotation. Based on ChineseEcomQA, we conduct extensive evaluations on mainstream LLMs and provide some valuable insights. We hope that ChineseEcomQA could guide future domain-specific evaluations, and facilitate broader LLM adoption in e-commerce applications.