14 papers
Manacá-1B-Instruct-2: Reproducible Safety Alignment for a Native Brazilian-Portuguese LLM. Negative DPO Results, a Controlled Composition Scan, and a Silent Failure Mode of the LLM Judge
Bruno Leonardo Santos Menezes, Carlos Leonardo Souza Cardoso, Fabio André Machado Porto
Brazilian Portuguese remains underserved by open-source language models, and the few that are safety-aligned rarely document what did not work or validate comparisons with statisti…
ZipTok3D: High-Fidelity 3D Tokenization with Compact Token Prefixes
Mingda Lin, Weijie Wang, Zeyu Zhang +7
Compact token sequences are essential for efficient 3D generation. However, existing 3D tokenizers typically organize latent representations either over spatial regions or as fixed…
Block3D: Efficient Text-to-3D Generation via Block-Wise Diffusion
Bowen Cui, Weijie Wang, Zeyu Zhang +7
While text-to-3D generation has advanced rapidly, achieving high geometric fidelity at low inference cost remains challenging. Existing text-to-3D methods either decode discrete sh…
LogiShot: Logically Coherent Cross-Shot Video Generation
Shuai Guo, Yuhang Yang, Zeyu Zhang +4
Generating cross-shot videos that are logically connected is essential for content creation. Currently, most cross-shot video-generation workflows, such as short-drama production,…
WorldOlympiad: Can Your World Model Survive a Triathlon?
Yuke Zhao, Wangbo Zhao, Weijie Wang +8
We introduce WorldOlympiad, a benchmark for diagnosing video-based world models across physical faithfulness, geometric consistency, and interaction fidelity. While existing benchm…
Latent Spatial Memory for Video World Models
Weijie Wang, Haoyu Zhao, Yifan Yang +7
Video world models that maintain 3D spatial consistency across generated frames typically rely on explicit point cloud memory constructed in RGB space. This design is both computat…
World-R1: Reinforcing 3D Constraints for Text-to-Video Generation
Weijie Wang, Xiaoxuan He, Youping Gu +9
Recent video foundation models demonstrate impressive visual synthesis but frequently suffer from geometric inconsistencies. While existing methods attempt to inject 3D priors via…
Less Detail, Better Answers: Degradation-Driven Prompting for VQA
Haoxuan Han, Weijie Wang, Zeyu Zhang +2
Recent advancements in Vision-Language Models (VLMs) have significantly pushed the boundaries of Visual Question Answering (VQA).However,high-resolution details can sometimes becom…
CoV: Chain-of-View Prompting for Spatial Reasoning
Haoyu Zhao, Akide Liu, Zeyu Zhang +5
Embodied question answering (EQA) in 3D environments often requires collecting context that is distributed across multiple viewpoints and partially occluded. However, most recent v…
DriveGen3D: Boosting Feed-Forward Driving Scene Generation with Efficient Video Diffusion
Weijie Wang, Jiagang Zhu, Zeyu Zhang +14
We present DriveGen3D, a novel framework for generating high-quality and highly controllable dynamic 3D driving scenes that addresses critical limitations in existing methodologies…
VolSplat: Rethinking Feed-Forward 3D Gaussian Splatting with Voxel-Aligned Prediction
Weijie Wang, Yeqing Chen, Zeyu Zhang +9
Feed-forward 3D Gaussian Splatting (3DGS) has emerged as a highly effective solution for novel view synthesis. Existing methods predominantly rely on a \emph{pixel-aligned} Gaussia…
FPSAttention: Training-Aware FP8 and Sparsity Co-Design for Fast Video Diffusion
Akide Liu, Zeyu Zhang, Zhexin Li +12
Diffusion generative models have become the standard for producing high-quality, coherent video content, yet their slow inference speeds and high computational demands hinder pract…
Revisiting Depth Representations for Feed-Forward 3D Gaussian Splatting
Duochao Shi, Weijie Wang, Donny Y. Chen +4
Depth maps are widely used in feed-forward 3D Gaussian Splatting (3DGS) pipelines by unprojecting them into 3D point clouds for novel view synthesis. This approach offers advantage…
ZPressor: Bottleneck-Aware Compression for Scalable Feed-Forward 3DGS
Weijie Wang, Donny Y. Chen, Zeyu Zhang +3
Feed-forward 3D Gaussian Splatting (3DGS) models have recently emerged as a promising solution for novel view synthesis, enabling one-pass inference without the need for per-scene…