5 papers
How to Utilize Complementary Vision-Text Information for 2D Structure Understanding
Jiancheng Dong, Pengyue Jia, Derong Xu +9
LLMs typically linearize 2D tables into 1D sequences to fit their autoregressive architecture, which weakens row-column adjacency and other layout cues. In contrast, purely visual…
MOSAIC: Composable Safety Alignment with Modular Control Tokens
Jingyu Peng, Hongyu Chen, Jiancheng Dong +5
Safety alignment in large language models (LLMs) is commonly implemented as a single static policy embedded in model parameters. However, real-world deployments often require conte…
SkyLink: A Large Vision-Language Model Driven Re-ranking Framework for Cross-View UAV geolocalization
Bowen Liu, Pengyue Jia, Wanyu Wang +9
Cross-view UAV geolocalization is fundamentally a challenging large-scale image retrieval task, aiming to determine the geographic coordinates of Unmanned Aerial Vehicle (UAV) quer…
Learning a Single Token to Replace Long System Prompts in LLMs
Jiancheng Dong, Pengyue Jia, Jingyu Peng +7
Long system prompts are widely used to steer Large Language Models (LLMs), but repeatedly processing them at inference time is inefficient and consumes valuable context budget. Thi…
Threshold Filtering Packing for Supervised Fine-Tuning: Training Related Samples within Packs
Jiancheng Dong, Lei Jiang, Wei Jin +1
Packing for Supervised Fine-Tuning (SFT) in autoregressive models involves concatenating data points of varying lengths until reaching the designed maximum length to facilitate GPU…