28 papers
PACER: Process-Aware VLA Policy Learning from Human Corrections for Robotic Disassembly
Chang Liu, Sibo Tian, Litian Gong +4
Robotic disassembly of end-of-life (EoL) products requires robots to accommodate product variation, follow prerequisite sequences, and approach components precisely for contact-ric…
Contextual Credit Policy Optimization
Huiwen Han, Ziyao Zeng, Zihao Zhu +6
Group-based reinforcement learning enables large language model agents to learn from sparse outcome rewards in long-horizon tasks. Recent methods derive step-level credit by compar…
CyberArena: Benchmarking AI Agents in Live Symmetric Attack-and-Defense Combat
Bangya Liu, Jianhong Tu, Yueyang Pan +18
AI agents are demonstrating increasingly strong offensive and defensive cybersecurity capabilities in the real world. Current benchmarks mostly evaluate agents in isolation against…
What Matters in Quality-Aware VLA Pretraining: A Benchmark and Empirical Study
Mingxuan Yan, Litian Gong, Dihong Huang +19
How should demonstration quality be used during vision-language-action (VLA) pretraining to improve performance after target-task fine-tuning? We compare weighting, filtering, and…
Training-Free Multi-Agent Cooperative Embodied Search from Uncertain Vision-Language Priors
Fan Pu, Keshu Wu, Jiwan Jiang +5
Real-world Vision-Language Navigation (VLN) requires agents to act under partial observability, where map structure, semantic cues, and target locations may be incomplete or incorr…
Disentangling Task-Relevant Latent Dynamics for Reward Modeling
Dayou Li, Jionghao Wang, Peihao Wang +9
Reward design remains a major bottleneck for reinforcement learning in robotic manipulation. Handcrafted dense rewards can compensate for the limited guidance provided by sparse re…
RoboIRS: Inference-Time Internal Representation Steering for Generalist Robot Policies
Jiuzhou Lei, Chang Liu, Dayou Li +5
Vision-language-action (VLA) and world-action models (WAMs) often degrade under out-of-distribution task variations despite retaining partial task capability. To recover such capab…
Dexterous Tactile World Model
Ziyao Zeng, Xiatao Sun, Hao Wang +6
World models for manipulation are typically trained from video, yet the events that determine how manipulation unfolds, such as making and releasing contact, are difficult to obser…
AnaDiffusion: Anatomically Compositional Latent Diffusion for Controllable 3D Brain MRI Generation
Huiwen Han, Lulin Liu, Bangya Liu +8
3D brain MRI generation has made significant advances in medical imaging, simulation, and controllable anatomical analysis. However, existing generative models typically synthesize…
EndoMD-SLAM: Endoscopic Gaussian Splatting SLAM under Optical Degradation with Memory and Static-Transient Decomposition
Nuo Chen, Kangqi Ni, Lulin Liu +5
Dense 3D reconstruction is critical for clinical endoscopic navigation and documentation. While Gaussian Splatting SLAM systems show promise in this domain, they fundamentally rely…
OpenLongTail: Generative Scaling of Long-Tail Driving Data
Lulin Liu, Nuo Chen, Yan Wang +15
Scaling robust driving policies is fundamentally bottlenecked by the scarcity of edge cases in curated datasets. While the real world continuously captures these critical events, s…
A Physics-Grounded Benchmark for Multi-Agent Dynamics in World Models
Nuo Chen, Lulin Liu, Zihao Li +12
Generative world models hold immense promise as scalable simulators for autonomous systems, particularly for synthesizing rare but safety-critical multi-agent interactions, such as…
DarkVGGT: Seeing Through Darkness Using Thermal Geometry
Minseong Kweon, Wenyuan Zhao, Nuo Chen +6
Recent feed-forward 3D reconstruction methods have demonstrated strong performance and flexibility in efficient end-to-end scene geometry estimation from image streams. However, th…
Behavior Uncloning: Distilling Mode Redirection into Policy Weights without Inference-Time Steering
Hao Wang, Jiuzhou Lei, Dayou Li +5
Behavior-cloned policies often learn multiple behavior modes from demonstration datasets, including modes that are unsafe or otherwise undesired at deployment. For example, a polic…
Embedding Semantic Risk into Distance Fields and CBFs for Online Monocular Safe Control
Dawei Zhang, Nuo Chen, Shuo Liu +2
We propose an online monocular perception-to-control framework that embeds semantic risk into the distance field used by Control Barrier Function (CBF)-based safe navigation and te…
HumanNOVA: Photorealistic, Universal and Rapid 3D Human Avatar Modeling from a Single Image
Hezhen Hu, Wangbo Zhao, Lanqing Guo +6
In this paper, we present HumanNOVA, a photorealistic, universal, and rapid model for generating 3D human avatars from a single RGB image. Achieving both photorealism and generaliz…
SpatialStack: Layered Geometry-Language Fusion for 3D VLM Spatial Reasoning
Jian Zhang, Shijie Zhou, Bangya Liu +2
Large vision-language models (VLMs) still struggle with reliable 3D spatial reasoning, a core capability for embodied and physical AI systems. This limitation arises from their ina…
Accelerating Transformer-Based Monocular SLAM via Geometric Utility Scoring
Xinmiao Xiong, Bangya Liu, Hao Wang +7
Geometric Foundation Models (GFMs) have recently advanced monocular SLAM by providing robust, calibration-free 3D priors. However, deploying these models on dense video streams int…
FF3R: Feedforward Feature 3D Reconstruction from Unconstrained views
Chaoyi Zhou, Run Wang, Feng Luo +4
Recent advances in vision foundation models have revolutionized geometry reconstruction and semantic understanding. Yet, most of the existing approaches treat these capabilities in…
EgoTL: Egocentric Think-Aloud Chains for Long-Horizon Tasks
Lulin Liu, Dayou Li, Yiqing Liang +8
Large foundation models have made significant advances in embodied intelligence, enabling synthesis and reasoning over egocentric input for household tasks. However, VLM-based auto…
A Statistical Framework for Auditing Behavioral Dependence and Induced Bias in LLM Judges
Chenchen Kuai, Jiwan Jiang, Zihao Zhu +8
The rapid growth of the large language model (LLM) ecosystem raises a critical question: are seemingly diverse models truly independent? Shared pretraining data, distillation, and…
Egocentric World Model for Photorealistic Hand-Object Interaction Synthesis
Dayou Li, Lulin Liu, Bangya Liu +6
To serve as a scalable data source for embodied AI, world models should act as true simulators that infer interaction dynamics strictly from user actions, rather than mere conditio…
Learning Actionable Manipulation Recovery via Counterfactual Failure Synthesis
Dayou Li, Jiuzhou Lei, Hao Wang +6
While recent foundation models have significantly advanced robotic manipulation, these systems still struggle to autonomously recover from execution errors. Current failure-learnin…
NavTrust: Benchmarking Trustworthiness for Embodied Navigation
Huaide Jiang, Yash Chaudhary, Yuping Wang +8
There are two major categories of embodied navigation: Vision-Language Navigation (VLN), where agents navigate by following natural language instructions; and Object-Goal Navigatio…