17 papers
EmbodiedEvo: Self-Evolving Coding Agent For Physical Intelligence
Wenyi Li, Yizhe Chi, Youjie Zheng +8
Coding agents can construct executable programs and revise them through feedback, offering a foundation for embodied control without target-task fine-tuning. Through a programmatic…
Language-Forcing: Advancing Humanoid Loco-Manipulation via Semantic Alignment of Whole-Body World Action Models
Yanjun Li, Jiarui Ye, Weiyu Guo +7
World Action Models (WAMs) jointly predict robot actions and future observations. However, their visual prediction objectives do not explicitly distinguish task-relevant transition…
SWE Refactor Bench: Can Coding Agents Complete a Long-Horizon, Whole-Repository Stack Migration?
Deyao Hong, Yizhe Chi, Wenyi Li +7
Modern software systems accumulate technical debt over decades of development, which makes migration expensive and largely manual. As coding agents become increasingly capable at b…
AI4AI-Bench: Benchmarking LLM Agents in Algorithmic Design for Recursive Self-Improvement
Yizhe Chi, Wenyi Li, Deyao Hong +7
Recursive self-improvement (RSI) asks whether an AI system can improve the process that produces AI systems, so that the next system inherits the improvement. That process is the t…
AdvFD: Boosting Visual Generation via Adversarial Fr'echet Distance Loss
Mingju Gao, Jingkai Zhou, Kun Gai +2
Fréchet distance has recently emerged as an effective distribution-level objective for generator post-training, complementing the conventional sample-level diffusion and flow-match…
Agogic: Performance-Timed Music Tokens for LLM-Native Text-to-Symbolic-Music Generation
Junhao Chen, Mingjin Chen, Jingjia Mao +12
Text-to-music language models begin with a choice usually made by default: how to tokenize music. Normally entangled with backbone, data, and recipe, its effect has never been meas…
PhysRAG: Enhancing Physics-Awareness in Video Generation via Retrieval-Augmented Generation
Kexu Cheng, Zicheng Liu, Mingju Gao +2
Developing physically aware video generation models remains a significant challenge due to the difficulty in capturing diverse physical phenomena, such as thermal dynamics, mechani…
SAID: Accelerating Diffusion-Based Language Models via Scaffold-Aware Iterative Decoding
Na Li, Chengda Wang, Mingju Gao +1
Diffusion large language models (DLLMs) enable non-autoregressive generation by iteratively denoising corrupted token sequences with bidirectional context. Despite their ability to…
SemBlock: Semantic Boundary Dynamic Blocks for Diffusion LLMs
Xinrui Song, Zhuoran Wang, Mingju Gao +1
Diffusion language models (DLMs) generate text through iterative denoising, and blockwise decoding improves their practicality by committing tokens in local blocks. However, existi…
PAM: A Pose-Appearance-Motion Engine for Sim-to-Real HOI Video Generation
Mingju Gao, Kaisen Yang, Huan-ang Gao +13
Hand-object interaction (HOI) reconstruction and synthesis are becoming central to embodied AI and AR/VR. Yet, despite rapid progress, existing HOI generation research remains frag…
Benchmarking PhD-Level Coding in 3D Geometric Computer Vision
Wenyi Li, Renkai Luo, Yue Yu +5
AI-assisted coding has rapidly reshaped software practice and research workflows, yet today's models still struggle to produce correct code for complex 3D geometric vision. If mode…
Challenger: Affordable Adversarial Driving Video Generation
Zhiyuan Xu, Bohan Li, Huan-ang Gao +7
Generating photorealistic driving videos has seen significant progress recently, but current methods largely focus on ordinary, non-adversarial scenarios. Meanwhile, efforts to gen…
FB-4D: Spatial-Temporal Coherent Dynamic 3D Content Generation with Feature Banks
Jinwei Li, Huan-ang Gao, Wenyi Li +13
With the rapid advancements in diffusion models and 3D generation techniques, dynamic 3D content generation has become a crucial research area. However, achieving high-fidelity 4D…
PartRM: Modeling Part-Level Dynamics with Large Cross-State Reconstruction Model
Mingju Gao, Yike Pan, Huan-ang Gao +6
As interest grows in world models that predict future states from current observations and actions, accurately modeling part-level dynamics has become increasingly relevant for var…
Training-Free Model Merging for Multi-target Domain Adaptation
Wenyi Li, Huan-ang Gao, Mingju Gao +3
In this paper, we study multi-target domain adaptation of scene understanding models. While previous methods achieved commendable results through inter-domain consistency losses, t…
FairDiff: Fair Segmentation with Point-Image Diffusion
Wenyi Li, Haoran Xu, Guiyu Zhang +4
Fairness is an important topic for medical image analysis, driven by the challenge of unbalanced training data among diverse target groups and the societal demand for equitable med…
SCP-Diff: Spatial-Categorical Joint Prior for Diffusion Based Semantic Image Synthesis
Huan-ang Gao, Mingju Gao, Jiaju Li +4
Semantic image synthesis (SIS) shows good promises for sensor simulation. However, current best practices in this field, based on GANs, have not yet reached the desired level of qu…