7 citations · 10 across the 27 of their papers we have counts for
21 papers · 1 filter
Decomposed On-Policy Distillation for Vision-Language Reasoning: Steering Gradients for Visual Grounding
Hee Suk Yoon, Eunseop Yoon, Jaehyun Jang +6
While on-policy distillation offers dense supervision for training small reasoning models, its optimization dynamics in the multimodal domain remain under-explored. In this work, w…
Lens: Rethinking Training Efficiency for Foundational Text-to-Image Models
Dong Chen, Fangyun Wei, Ziyu Wan +18
We introduce Lens, a 3.8B-parameter T2I model that achieves performance competitive with, and in several cases surpassing, state-of-the-art models with more than 6B parameters acro…
MM-WebAgent: A Hierarchical Multimodal Web Agent for Webpage Generation
Yan Li, Zezi Zeng, Yifan Yang +12
The rapid progress of Artificial Intelligence Generated Content (AIGC) tools enables images, videos, and visualizations to be created on demand for webpage design, offering a flexi…
High-Fidelity Text-to-Image Generation from Pre-Trained Vision-Language Models via Distribution-Conditioned Diffusion Decoding
Ji Woo Hong, Hee Suk Yoon, Gwanhyeong Koo +5
Recent large-scale vision-language models (VLMs) have shown remarkable text-to-image generation capabilities, yet their visual fidelity remains constrained by the discrete image to…
Video-in-the-Loop: Span-Grounded Long Video QA with Interleaved Reasoning
Chendong Wang, Donglin Bai, Yifan Yang +11
We present \emph{Video-in-the-Loop} (ViTL), a two-stage long-video QA framework that preserves a fixed token budget by first \emph{localizing} question-relevant interval(s) with a…
StableAvatar: Infinite-Length Audio-Driven Avatar Video Generation
Shuyuan Tu, Yueming Pan, Yinming Huang +6
Current diffusion models for audio-driven avatar video generation struggle to synthesize long videos with natural audio synchronization and identity consistency. This paper present…
Phi-Ground Tech Report: Advancing Perception in GUI Grounding
Miaosen Zhang, Ziqiang Xu, Jialiang Zhu +8
With the development of multimodal reasoning models, Computer Use Agents (CUAs), akin to Jarvis from \textit{"Iron Man"}, are becoming a reality. GUI grounding is a core component…
StableAnimator++: Overcoming Pose Misalignment and Face Distortion for Human Image Animation
Shuyuan Tu, Zhen Xing, Xintong Han +5
Current diffusion models for human image animation often struggle to maintain identity (ID) consistency, especially when the reference image and driving video differ significantly…
ReasonGen-R1: CoT for Autoregressive Image generation models through SFT and RL
Yu Zhang, Yunqi Li, Yifan Yang +7
Although chain-of-thought reasoning and reinforcement learning (RL) have driven breakthroughs in NLP, their integration into generative vision models remains underexplored. We intr…
ViaRL: Adaptive Temporal Grounding via Visual Iterated Amplification Reinforcement Learning
Ziqiang Xu, Qi Dai, Tian Xie +5
Video understanding is inherently intention-driven-humans naturally focus on relevant frames based on their goals. Recent advancements in multimodal large language models (MLLMs) h…
JointDiT: Enhancing RGB-Depth Joint Modeling with Diffusion Transformers
Kwon Byung-Ki, Qi Dai, Lee Hyoseok +2
We present JointDiT, a diffusion transformer that models the joint distribution of RGB and depth. By leveraging the architectural benefit and outstanding image prior of the state-o…
Learning Zero-Shot Subject-Driven Video Generation Using 1% Compute
Daneul Kim, Jingxu Zhang, Wonjoon Jin +4
Subject-driven video generation (SDV-Gen) aims to produce videos of a specific subject by adapting a pretrained video model, enabling personalized and application-driven content cr…
FloVD: Optical Flow Meets Video Diffusion Model for Enhanced Camera-Controlled Video Synthesis
Wonjoon Jin, Qi Dai, Chong Luo +2
We present FloVD, a novel video diffusion model for camera-controllable video generation. FloVD leverages optical flow to represent the motions of the camera and moving objects. Th…
HiTVideo: Hierarchical Tokenizers for Enhancing Text-to-Video Generation with Autoregressive Large Language Models
Ziqin Zhou, Yifan Yang, Yuqing Yang +7
Text-to-video generation poses significant challenges due to the inherent complexity of video data, which spans both temporal and spatial dimensions. It introduces additional redun…
MageBench: Bridging Large Multimodal Models to Agents
Miaosen Zhang, Qi Dai, Yifan Yang +6
LMMs have shown impressive visual understanding capabilities, with the potential to be applied in agents, which demand strong reasoning and planning abilities. Nevertheless, existi…
StableAnimator: High-Quality Identity-Preserving Human Image Animation
Shuyuan Tu, Zhen Xing, Xintong Han +4
Current diffusion models for human image animation struggle to ensure identity (ID) consistency. This paper presents StableAnimator, the first end-to-end ID-preserving video diffus…
REDUCIO! Generating 1K Video within 16 Seconds using Extremely Compressed Motion Latents
Rui Tian, Qi Dai, Jianmin Bao +5
Commercial video generation models have exhibited realistic, high-fidelity results but are still restricted to limited access. One crucial obstacle for large-scale applications is…
LLM2CLIP: Powerful Language Model Unlocks Richer Cross-Modality Representation
Weiquan Huang, Aoqi Wu, Yifan Yang +10
CLIP is a seminal multimodal model that maps images and text into a shared representation space through contrastive learning on billions of image-caption pairs. Inspired by the rap…
Aligning Vision Models with Human Aesthetics in Retrieval: Benchmarks and Algorithms
Miaosen Zhang, Yixuan Wei, Zhen Xing +8
Modern vision models are trained on very large noisy datasets. While these models acquire strong capabilities, they may not follow the user's intent to output the desired results i…
MotionFollower: Editing Video Motion via Lightweight Score-Guided Diffusion
Shuyuan Tu, Qi Dai, Zihao Zhang +6
Despite impressive advancements in diffusion-based video editing models in altering video attributes, there has been limited exploration into modifying motion information while pre…
OmniVid: A Generative Framework for Universal Video Understanding
Junke Wang, Dongdong Chen, Chong Luo +4
The core of video understanding tasks, such as recognition, captioning, and tracking, is to automatically detect objects or actions in a video and analyze their temporal evolution.…