Search Results for author:
Found 6 papers, 3 papers with code
Towards Building a Robust Knowledge Intensive Question Answering Model with Large Language Models
The development of LLMs has greatly enhanced the intelligence and fluency of question answering, while the emergence of retrieval enhancement has enabled models to better utilize external information.
Large Language Models Understand Layout
The experimental results reveal that the layout understanding ability of LLMs is mainly introduced by the coding data for pretraining, which is further enhanced at the instruction-tuning stage.
Revisiting the Power of Prompt for Visual Tuning
Inspired by the observation that the prompt tokens tend to share high mutual information with patch tokens, we propose initializing prompts with downstream token prototypes.
Ranked #1 on
Visual Prompt Tuning
on VTAB-1k(Structured<8>)
Improving Knowledge Distillation via Regularizing Feature Norm and Direction
Finally, we propose a rather simple loss term (dubbed ND loss) to simultaneously (1) encourage student to produce large-\emph{norm} features, and (2) align the \emph{direction} of student features and teacher class-means.
Ranked #1 on
Knowledge Distillation
on ImageNet
FocalClick: Towards Practical Interactive Image Segmentation
To make the model work with preexisting masks, we formulate a sub-task termed Interactive Mask Correction, and propose Progressive Merge as the solution.
Ranked #2 on
Interactive Segmentation
on DAVIS
(using extra training data)
Conditional Diffusion for Interactive Segmentation
In click-based interactive segmentation, the mask extraction process is dictated by positive/negative user clicks; however, most existing methods do not fully exploit the user cues, requiring excessive numbers of clicks for satisfactory results.
