14 papers
Dyn-3D: Unveiling and Resolving Ego-Motion Ambiguity in Vision-Language Models
Jiayu Ding, Zhuodong Liu, Lei Zhang +6
As Vision-Language Models (VLMs) tackle dynamic 3D spatial reasoning, ego-motion perception becomes essential to resolve monocular scale ambiguity. However, current models often ov…
AdaThinkV: Adaptive Thinking for Token-Efficient Video Reasoning
Jingqi Tian, Haoji Zhang, Lin Chen +7
Chain-of-thought (CoT) reasoning can improve performance on difficult video questions but often wastes decoding tokens on simple ones. We study whether a video multimodal large lan…
ZeroSplat: Generalized Referring Segmentation in 3D Gaussian Splatting
Jiayu Ding, Meilu Song, Xiaoyi Zhang +3
Recent advancements in 3D Gaussian Splatting (3DGS) have enabled language-guided scene understanding. However, existing Referring 3D Gaussian Splatting (R3DGS) methods are fundamen…
ContextGuard: Structured Self-Auditing for Context Learning in Language Models
Hongbo Jin, Chi Wang, Haoran Tang +5
Recent benchmarks reveal that despite strong reasoning capabilities, large language models (LLMs) still struggle to faithfully apply complex contextual knowledge. These failures ar…
Context-CoT: Enhancing Context Learning via High-Quality Reasoning Synthesis
Hongbo Jin, Mingnan Zhu, Jingqi Tian +6
While LLMs excel at reasoning over prompts using static pretrained knowledge, they struggle significantly with context learning-the ability to dynamically extract, internalize, and…
DGPO: Distribution Guided Policy Optimization for Fine Grained Credit Assignment
Hongbo Jin, Rongpeng Zhu, Zhongjing Du +4
Reinforcement learning is crucial for aligning large language models to perform complex reasoning tasks. However, current algorithms such as Group Relative Policy Optimization suff…
VISD: Enhancing Video Reasoning via Structured Self-Distillation
Hao Lin, Kunyang Lv, Xu Jiang +5
Training VideoLLMs for complex reasoning remains challenging due to sparse sequence level rewards and the lack of fine grained credit assignment over long, temporally grounded reas…
HiMAC: Hierarchical Macro-Micro Learning for Long-Horizon LLM Agents
Hongbo Jin, Rongpeng Zhu, Jiayu Ding +2
Large language model (LLM) agents have recently demonstrated strong capabilities in interactive decision-making, yet they remain fundamentally limited in long-horizon tasks that re…
TIR-Flow: Active Video Search and Reasoning with Frozen VLMs
Hongbo Jin, Siyi Xie, Jiayu Ding +2
While Large Video-Language Models (Video-LLMs) have achieved remarkable progress in perception, their reasoning capabilities remain a bottleneck. Existing solutions typically resor…
3D Instruction Ambiguity Detection
Jiayu Ding, Haoran Tang, Hongbo Jin +2
In safety-critical domains, linguistic ambiguity can have severe consequences; a vague command like "Pass me the vial" in a surgical setting could lead to catastrophic errors. Yet,…
VideoCuRL: Video Curriculum Reinforcement Learning with Orthogonal Difficulty Decomposition
Hongbo Jin, Kuanwei Lin, Wenhao Zhang +2
Reinforcement Learning (RL) is crucial for empowering VideoLLMs with complex spatiotemporal reasoning. However, current RL paradigms predominantly rely on random data shuffling or…
VideoMem: Enhancing Ultra-Long Video Understanding via Adaptive Memory Management
Hongbo Jin, Qingyuan Wang, Wenhao Zhang +2
Ultra long video understanding remains an open challenge, as existing vision language models (VLMs) falter on such content due to limited context length and inefficient long term m…
Enhancing Long Video Question Answering with Scene-Localized Frame Grouping
Xuyi Yang, Wenhao Zhang, Hongbo Jin +5
Current Multimodal Large Language Models (MLLMs) often perform poorly in long video understanding, primarily due to resource limitations that prevent them from processing all video…
VISTA: Mitigating Semantic Inertia in Video-LLMs via Training-Free Dynamic Chain-of-Thought Routing
Hongbo Jin, Jiayu Ding, Siyi Xie +2
Recent advancements in Large Language Models have successfully transitioned towards System 2 reasoning, yet applying these paradigms to video understanding remains challenging. Whi…