Parameter-efficient fine-tuning (PEFT) is a standard way to adapt multimodal large language models, yet extremely low-rank settings -- especially rank-1 LoRA -- are often unstable. We show that this instability is not solely due to limited capacity: in the rank-1 regime, optimization is highly sensitive to the update direction. Concretely, pretrained vision and text features form mismatched anisotropic regions, yielding a dominant "gap" direction that acts like a translation component and disproportionately steers early gradients under rank-1 constraints. Analyzing pretrained representations, we identify a modality-gap axis that dominates early gradient flow, while a random rank-1 initialization is unlikely to align with it, leading to weak gradients and training collapse. We propose Gap-Init, a geometry-aware initialization that aligns the rank-1 LoRA direction with an estimated modality-gap vector from a small calibration set, while keeping the initial LoRA update zero. Across multiple vision-language tasks and backbones, Gap-Init consistently stabilizes rank-1 training and can match or outperform strong rank-8 baselines. Our results suggest that at the extreme low-rank limit, initial alignment can matter as much as rank itself.
Learning in the Fisher Subspace: A Guided Initialization for LoRA Fine-Tuning
Department of Computer Science（计算机科学系）；Information Engineering, National Cheng Kung University（信息工程系，国立成功大学）；Department of Artificial Intelligence, Chang Gung University（人工智能系，长庚大学）；Department of Computer Science, National Tsing Hua University（计算机科学系，国立清华大学）
Improving Fast Adversarial Training with Prior-Guided Knowledge
Institute of Information Engineering, Chinese Academy of Sciences（中国科学院信息工程研究所）；University of Chinese Academy of Sciences（中国科学院大学）；Tencent Inc.（腾讯公司）；Beihang University（北京航空航天大学）；The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））；Shenzhen Research Institute of Big Data（深圳大数据研究院）；Sun Yat-sen University（中山大学）
LLM-Guided Initialization for Accelerated Hybrid Quantum-Classical Medical Image Classification
Faculty of Computing Universiti Teknologi Malaysia Johor Bahru, Malaysia Remote Student based in Jakarta, Indonesia；Informatics Department Faculty of Science；Faculty of Computing Universiti Teknologi Malaysia Johor Bahru, Malaysia
Teacher-Student Gaps Are Not Enough: Outcome-Guided On-Policy Distillation for Multi-Turn Autonomous Agents
Qwen Large Model Application Team, Alibaba（阿里巴巴通义千问大模型应用团队）；Peking University（北京大学）；University of Chinese Academy of Sciences（中国科学院大学）；Shanghai Jiao Tong University（上海交通大学）
Learning High-Degree Parities: The Crucial Role of the Initialization
Ecole Polytechnique Fédérale de Lausanne (EPFL)（洛桑联邦理工学院）；Massachusetts Institute of Technology (MIT)（麻省理工学院）；Inria Paris（法国国家信息与自动化研究所巴黎分部）；PSL University（巴黎文理研究大学）；African Institute for Mathematical Sciences (AIMS)（非洲数学科学研究院）
Alignment-Aware Model Adaptation via Feedback-Guided Optimization
Department of Computer Science, University of British Columbia, Location, Canada（不列颠哥伦比亚大学计算机科学系）；Vector Institute of AI, Toronto, Canada（人工智能向量研究所）；Stony Brook University, New York, USA（石溪大学）
Enhancing Pretrained Model-based Continual Representation Learning via Guided Random Projection
Wuhan University（武汉大学）；Shanghai Innovation Institute（上海创新研究院）；Shanghai Jiao Tong University（上海交通大学）；Tsinghua University（清华大学）；University of Science and Technology of China（中国科学技术大学）；China University of Mining and Technology（中国矿业大学）
