Xin Jin's picture

Xin Jin

Xin1118

·

https://github.com/JinXins

AI & ML interests

None yet

Recent Activity

upvoted a paper 5 days ago

EarlyTom: Early Token Compression Completes Fast Video Understanding

authored a paper 9 days ago

MergeMix: A Unified Augmentation Paradigm for Visual and Multi-Modal Understanding

authored a paper 9 days ago

LVOmniBench: Pioneering Long Audio-Video Understanding Evaluation for Omnimodal LLMs

View all activity

Organizations

None yet

authored 3 papers 9 days ago

MergeMix: A Unified Augmentation Paradigm for Visual and Multi-Modal Understanding

Paper • 2510.23479 • Published Oct 27, 2025 • 18

LVOmniBench: Pioneering Long Audio-Video Understanding Evaluation for Omnimodal LLMs

Paper • 2603.19217 • Published Mar 19 • 28

RankE: End-to-End Post-Training for Discrete Text-to-Image Generation with Decoder Co-Evolution

Paper • 2605.21195 • Published 14 days ago • 19

authored 4 papers 12 months ago

Adversarial AutoMixup

Paper • 2312.11954 • Published Dec 19, 2023 • 4

A Survey on Mixup Augmentations and Beyond

Paper • 2409.05202 • Published Sep 8, 2024 • 1

SUMix: Mixup with Semantic and Uncertain Information

Paper • 2407.07805 • Published Jul 10, 2024

Taming LLMs by Scaling Learning Rates with Gradient Grouping

Paper • 2506.01049 • Published Jun 1, 2025 • 39