9 papers
ReSAIL: Mitigating Collapse in Iterative Agent Self-Distillation
Shengjie Jin, Hengbo Xu, Zelong Sun +2
Iterative self-distillation enables LLM agents to learn from successive deployments, offering a path toward recursive self-improvement (RSI). Yet our experiments with existing meth…
Learning from Failures: Retrieval-Centric CoT via Hard Negatives for Unified Multimodal Retrieval
Zelong Sun, Jun Wang, Kaicheng Yang +3
Unified multimodal retrieval aims to identify candidates that satisfy complex user intent expressed through heterogeneous inputs. Although Large Vision-Language Model (LVLM)-based…
Learning Action Priors for Cross-embodiment Robot Manipulation
Dong Jing, Tianqi Zhang, Jiaqi Liu +5
Most Vision-Language-Action (VLA) models build on a Vision-Language Model (VLM) backbone by attaching an action module and optimizing the full policy jointly. This design inherits…
UniDoc-RL: Coarse-to-Fine Visual RAG with Hierarchical Actions and Dense Rewards
Jun Wang, Shuo Tan, Zelong Sun +5
Retrieval-Augmented Generation (RAG) extends Large Vision-Language Models (LVLMs) with external visual knowledge. However, existing visual RAG systems typically rely on generic ret…
Say Cheese! Detail-Preserving Portrait Collection Generation via Natural Language Edits
Zelong Sun, Jiahui Wu, Ying Ba +2
As social media platforms proliferate, users increasingly demand intuitive ways to create diverse, high-quality portrait collections. In this work, we introduce Portrait Collection…
DanQing: An Up-to-Date Large-Scale Chinese Vision-Language Pre-training Dataset
Hengyu Shen, Tiancheng Gu, Bin Qin +10
Vision-Language Pre-training (VLP) models have achieved remarkable success by leveraging large-scale image-text pairs. While English-centric models like CLIP and SigLIP benefit fro…
Mixture of Horizons in Action Chunking
Dong Jing, Gang Wang, Jiaqi Liu +7
Vision-language-action (VLA) models have shown remarkable capabilities in robotic manipulation, but their performance is sensitive to the used during…
CoTMR: Chain-of-Thought Multi-Scale Reasoning for Training-Free Zero-Shot Composed Image Retrieval
Zelong Sun, Dong Jing, Zhiwu Lu
Zero-Shot Composed Image Retrieval (ZS-CIR) aims to retrieve target images by integrating information from a composed query (reference image and modification text) without training…
Leveraging Large Vision-Language Model as User Intent-aware Encoder for Composed Image Retrieval
Zelong Sun, Dong Jing, Guoxing Yang +2
Composed Image Retrieval (CIR) aims to retrieve target images from candidate set using a hybrid-modality query consisting of a reference image and a relative caption that describes…