paper-with-me

홈 › Papers

Knowledge-Preserved Model Tuning in Null-Space for Robust Spatio-Temporal Video Grounding

2026-06-02 · Haoxuan Chen, Xianqin Liu, Jian-Fang Hu arxiv

Spatio-Temporal Video Grounding aims to localize object tubes based on textual queries. While recent methods have achieved remarkable success, they mainly focus on high-quality(HQ) inputs, neglecting the widespread presence of low-quality(LQ) videos in real-world scenarios. Although tuning methods like LoRA can adapt to degraded inputs, they inevitably disrupt pre-trained knowledge. To address this, we propose Null-Space Tuning (NST). This framework exploits the geometric property that adding vectors within the null-space of frozen weights to the layer input does not affect the output. Leveraging this, NST injects learnable residuals into input features that can be selectively invisible to the pre-trained backbone. Specifically, NST combines the Quality-Adaptive Unit and Dual-Space Reparameterization to synthesize these residuals by confining components for HQ inputs to the null-space, while directing restoration components for LQ inputs to the non-null space. As the frozen weights eliminate null-space components, we effectively rectify degraded inputs while preserving pre-trained knowledge for HQ inputs. Extensive experiments show that NST outperforms state-of-the-art methods on our Mixed-Quality benchmark.

📄 PDF Abstract BibTeX arXiv:2606.03539

Code (0)

등록된 구현이 없습니다.

Tasks

Spatio-Temporal Video Grounding

Similar Papers 제목 키워드 기반

AlphaEdit: Null-Space Constrained Knowledge Editing for Language Models

2024-10-03 · Junfeng Fang, Houcheng Jiang, Kun Wang, Yunshan Ma 외

Large language models (LLMs) often exhibit hallucinations due to incorrect or outdated knowledge. Hence, model editing methods have emerged to enable targeted knowledge updates. To achieve this, a prevailing paradigm is …

knowledge editingModel Editing

SonoEdit: Null-Space Constrained Knowledge Editing for Pronunciation Correction in LLM-Based TTS

2026-01-23 · Ayush Pratap Singh, Harshit Singh, Nityanand Mathur, Akshat Mandloi 외 arxiv

Neural text-to-speech (TTS) systems systematically mispronounce low-resource proper nouns, particularly non-English names, brands, and geographic locations, due to their underrepresentation in predominantly English train…

knowledge editing

EvoEdit: Evolving Null-space Alignment for Robust and Efficient Knowledge Editing

2025-10-11 · Sicheng Lyu, Yu Gu, Xinyu Wang, Jerry Huang 외 arxiv

Large language models (LLMs) require continual updates to rectify outdated or erroneous knowledge. Model editing has emerged as a compelling paradigm for introducing targeted modifications without the computational burde…

knowledge editing

SC-LoRA: Balancing Efficient Fine-tuning and Knowledge Preservation via Subspace-Constrained LoRA

2025-05-29 · Minrui Luo, Fuhang Kuang, Yu Wang, Zirui Liu 외

Parameter-Efficient Fine-Tuning (PEFT) methods, particularly Low-Rank Adaptation (LoRA), are indispensable for efficiently customizing Large Language Models (LLMs). However, vanilla LoRA suffers from slow convergence spe…

Navigateparameter-efficient fine-tuningWorld Knowledge

Approximate Nullspace Augmented Finetuning for Robust Vision Transformers

2024-03-15 · Haoyang Liu, Aditya Singh, Yijiang Li, Haohan Wang

Enhancing the robustness of deep learning models, particularly in the realm of vision transformers (ViTs), is crucial for their real-world deployment. In this work, we provide a finetuning approach to enhance the robustn…