Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileAcademic lineage
Possible advisorsa guess from early papers, not confirmed
- Mac SchwagerSuggested from co-authorship
- Suggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works68 from public data
- 1,232
- Foundation models in robotics: Applications, challenges, and the future459
How foundation models contribute to improving robot capabilities in the domains of perception, decision-making, and control is explored and the challenges hindering the adoption of foundation models in robot autonomy are discussed.
- MimicPlay: Long-Horizon Imitation Learning by Watching Human Play391
A hierarchical learning framework named MimicPlay is introduced that learns latent plans from human play data to guide low-level visuomotor control trained on a small number of teleoperated demonstrations and outperforms state-of-the-art imitation learning methods in task success rate, generalization ability, and robustness to disturbances.
- Cross-View Semantic Segmentation for Sensing Surroundings336
A novel visual task called Cross-view Semantic Segmentation as well as a framework named View Parsing Network (VPN) to address it and the experimental results show that the model can effectively make use of the information from different views and multi-modalities to understanding spatial information.
- Large AI Models in Health Informatics: Applications, Challenges, and the Future233
Seven key sectors in which large AI models are applicable and might have substantial influence, including: 1) bioinformatics; 2) medical diagnosis; 3) medical imaging; 4) medical informatics; 5) medical education; 6) public health; and 7) medical robotics are identified.
- BEHAVIOR-1K: A Human-Centered, Embodied AI Benchmark with 1,000 Everyday Activities and Realistic Simulation197
BEHAVIOR-1K's human-grounded nature, diversity, and realism make it valuable for embodied AI and robot learning research, and it is hoped that its human-grounded nature, diversity, and realism make it valuable for embodied AI and robot learning research.
- 176
- A Survey of Reasoning with Foundation Models: Concepts, Methodologies, and Outlook142
This article introduces seminal foundation models proposed or adaptable for reasoning, highlighting the latest advancements in various reasoning tasks, methods, and benchmarks, and delves into the potential future directions behind the emergence of reasoning abilities within foundation models.
- 100
- Dojo: A Differentiable Physics Engine for Robotics84
A differentiable rigid-body-dynamics simulator for robotics that prioritizes physical accuracy and differentiability is presented and the usefulness of the simulator and its gradients is demonstrated through a number of examples including: simulation, trajectory optimization, reinforcement learning, and system identification.
- Conformal Prediction for Uncertainty-Aware Planning with Diffusion Dynamics Model81
This paper quantifies the uncertainty of diffusion dynamics models using Conformal Prediction (CP), inspired by concepts from conformal prediction, and modify the loss function for training the diffusion model to include a quantile term to encourage more robust performance across the variety of training examples.
- A Survey of Reasoning with Foundation Models73
This paper introduces seminal foundation models proposed or adaptable for reasoning, highlighting the latest advancements in various reasoning tasks, methods, and benchmarks and discusses the relevance of multimodal learning, autonomous agents, and super alignment in the context of reasoning.
- PlaTe: Visually-Grounded Planning With Transformers in Procedural Tasks69
This work addresses the problem of how to leverage instructional videos to facilitate the understanding of human decision-making processes, focusing on training a model with the ability to plan a goal-directed procedure from real-world videos with Planning Transformer (PlaTe), which has the advantage of circumventing the compounding prediction errors that occur with single-step models during long model-based rollouts.
- SegVoxelNet: Exploring Semantic Context and Depth-aware Features for 3D Vehicle Detection from Point Cloud65
A semantic context encoder is proposed to leverage the free-of-charge semantic segmentation masks in the bird's eye view to address the semantic context information in LiDAR.
- 37
- ClawsBench: Evaluating Capability and Safety of LLM Productivity Agents in Simulated Workspaces36
ClawsBench, a benchmark for evaluating and improving LLM agents in realistic productivity settings, is introduced and eight recurring patterns of unsafe behavior are identified, including multi-step sandbox escalation and silent contract modification.
- Lyra: Orchestrating Dual Correction in Automated Theorem Proving35
The Lyra is introduced, a new framework that employs two distinct correction mechanisms: Tool Correction (TC) and Conjecture Correction (CC), an error feedback mechanism designed to interact with prover to refine formal proof conjectures with prover error messages.
- ParticleFormer: A 3D Point Cloud World Model for Multi-Object, Multi-Material Robotic Manipulation34
This work presents ParticleFormer, a Transformer-based point cloud world model trained with a hybrid point cloud reconstruction loss, supervising both global and local dynamics features in multi-material, multi-object robot interactions, and extends existing dynamics learning benchmarks to include diverse multi-material, multi-object interaction scenarios.
- Egocentric Human Trajectory Forecasting With a Wearable Camera and Multi-Modal Fusion29
A Transformer-based encoder-decoder neural network model, integrated with a novel cascaded cross-attention mechanism that fuses multiple modalities, has been designed to predict the future trajectory of the camera wearer, with results showing that the model outperforms the state-of-the-art methods in egocentric human trajectory forecasting.
- ARCH: Hierarchical Hybrid Learning for Long-Horizon Contact-Rich Robotic Assembly27
This work proposes a hierarchical modular approach, named Adaptive Robotic Compositional Hierarchy (ARCH), which enables long-horizon, high-precision robotic assembly in contact-rich settings and shows that ARCH generalizes well to unseen objects and outperforms baseline methods in terms of success rate and data efficiency.
- GRAD-NAV++: Vision-Language Model Enabled Visual Drone Nav igation With G aussian Ra diance Fields and D ifferentiable Dynamics26
This work introduces a lightweight Vision–Language–Action (VLA) framework that runs fully onboard and follows natural-language commands in real time and establishes a new benchmark for fully onboard Vision-Language-Action (VLA) flight and demonstrates that compact, efficient models can enable reliable, language-guided navigation without relying on external infrastructure.
- Transferable Active Grasping and Real Embodied Dataset23
A novel mask-guided reward is proposed to overcome the sparse reward issue in grasping and ensure category-irrelevant behavior in grasping in a real-world UR-5 robotic arm.
- Leveraging Print Debugging to Improve Code Generation in Large Language Models22
This work proposes an in-context learning approach that guides LLMs to debug by using a "print debugging" method, which involves inserting print statements to trace and analysing logs for fixing the bug.
- 22
- Research Challenges and Progress in the End-to-End V2X Cooperative Autonomous Driving Competition17
The design and outcomes of the End-to-End Autonomous Driving through V2X Cooperation Challenge are described, key research problems including bandwidth-aware fusion, robust multi-agent planning, and heterogeneous sensor integration are highlighted, and emerging technical trends among top-performing solutions are analyzed.
- NeRF-Loc: Transformer-Based Object Localization Within Neural Radiance Fields16
A transformer-based framework, NeRF-Loc, to extract 3D bounding boxes of objects in NeRF scenes, which is better than the baselines and compared with conventional RGB(-D) based methods that take rendered RGB images and depths from NeRFs as inputs.
- SIREN: Semantic, Initialization-Free Registration of Multi-Robot Gaussian Splatting Maps12
The superior performance of SIREN is demonstrated compared to competing baselines across a range of real-world datasets, and in particular, across the most widely-used robot hardware platforms, including a manipulator, drone, and quadruped.
- FAST-Splat: Fast, Ambiguity-Free Semantics Transfer in Gaussian Splatting12
FAST-Splat takes a bottom-up approach in deriving FAST-Splat, dismantling the limitations of closed-set semantic distillation to enable open-set (open-vocabulary) semantic distillation, and preserves the training, rendering, and memory-usage advantages of Gaussian Splatting over neural field methods.
- $π$, But Make It Fly: Physics-Guided Transfer of VLA Models to Aerial Manipulation11
AirVLA is introduced, a system that investigates the transferability of manipulation-pretrained VLAs to aerial pick-and-place tasks and a Payload-Aware Guidance mechanism that injects payload constraints directly into the policy's flow-matching sampling process increases real-world pick-and-place task success.
- Emerging Cyber Attack Risks of Medical AI Agents11
It is revealed that through adversarial prompts embedded on webpages, cyberattackers can inject false information into the agent's response and force the agent to manipulate recommendation, resulting in the leak of sensitive/private medical information.
- GRaD-Nav: Efficiently Learning Visual Drone Navigation with Gaussian Radiance Fields and Differentiable Dynamics11
A novel framework that integrates 3D Gaussian Splatting with differentiable deep reinforcement learning (DDRL) to train vision-based drone navigation policies and incorporates a Context-aided Estimator Network (CENet) to adapt to environmental variations at runtime.
- 11
- Breaking Lock-In: Preserving Steerability under Low-Data VLA Post-Training9
DeLock mitigates lock-in by preserving visual grounding during post-training and applying test-time contrastive prompt guidance to steer the policy's denoising dynamics according to novel instructions.
- SteerVLA: Steering Vision-Language-Action Models in Long-Tail Driving Scenarios9
SteerVLA is proposed, which leverages the reasoning capabilities of VLMs to produce fine-grained language instructions that steer a vision-language-action (VLA) driving policy, and outperforms state-of-the-art methods on a challenging closed-loop benchmark.
- Localization and recognition of human action in 3D using transformers9
This work introduces a new, challenging, and more complex benchmark dataset, BABEL-TAL (BT), for 3D action localization, and proposes a strong baseline model, i.e., Localizing Actions with Transformers (LocATe), that jointly localizes and recognizes actions in a 3D sequence.
- LocATe: End-to-end Localization of Actions in 3D with Transformers9
LocATe is proposed, an end-to-end approach that jointly localizes and recognizes actions in a 3D sequence and outperforms previous approaches on the existing PKU-MMD 3D-TAL benchmark (mAP=93.2%).
- 9
- SARM2: Multi-Task Stage Aware Reward Modeling for Self Improving Robotic Manipulation8
RM is introduced, a multi-task stage-aware reward model that combines an action-primitive-based stage estimator with a multi-gate Mixture-of-Experts value head to produce dense per-step rewards across manipulation tasks, showing that high-quality dense rewards are key to a stable robot data flywheel.
- 8
- Spatial RoboGrasp: Generalized Robotic Grasping Control Policy7
This paper proposes a unified framework that couples robust multimodal perception with reliable grasp prediction and demonstrates that spatially grounded perception, paired with diffusion-based imitation learning, offers a scalable and robust solution for general-purpose robotic grasping.
- Aria-NeRF: Multimodal Egocentric View Synthesis5
A comprehensive multimodal egocentric video dataset, featuring RGB images, eye-tracking camera footage, audio recordings from a microphone, atmospheric pressure readings from a barometer, positional coordinates from GPS, connectivity details from Wi-Fi and Bluetooth, and information from dual-frequency IMU datasets paired with a magnetometer are presented.
- Self-Supervised Traffic Advisors: Distributed, Multi-view Traffic Prediction for Smart Cities5
The present work proposes the Self-Supervised Traffic Advisor (SSTA), an infrastructure edge device concept that leverages self-supervised video prediction in concert with a communication and co-training framework to enable autonomously predicting traffic throughout a smart city.
- Unveiling the Entropy Dynamics of Chain-of-Thought Reasoning4
This paper investigates the entropy dynamics of Chain-of-Thought (CoT) and uncovers a consistent two-phase structure: an Uncertainty Region of exploration transitioning sharply to a Confidence Region of convergence, and develops a training-free framework for real-time inference control.
- Connected Autonomous Vehicle Motion Planning with Video Predictions from Smart, Self-Supervised Infrastructure4
This work leverages a recently-proposed “Self-Supervised Traffic Advisor” (SSTA) framework of smart sensors that teach themselves to generate and broadcast useful video predictions of road users that can effectively aid CAV motion planning.
- HiABP: Hierarchical Initialized ABP for Unsupervised Representation Learning4
Hierarchical Initialized Alternating Backpropagation (HiABP) is proposed for efficient Bayesian inference of hierarchical latent representation with controllable image synthesis and can outperform other popular deep generative models in modeling natural images and learning from incomplete data.
- 3
- Benchmarking the Thinking Mode of Multimodal Large Language Models in Clinical Tasks3
Evaluating the active"thinking mode"capabilities of two leading MLLMs, Seed1.5-VL and Gemini-2.5-Flash, for medical applications reveals that the improvement from activating the thinking mode remains marginal compared to the standard non-thinking mode for the majority of the tasks.
- Learning Driving Decisions by Imitating Drivers' Control Behaviors3
This work proposes a hybrid framework for learning a decision module, which is agnostic to the mechanisms of perception, planning, and control modules, and demonstrates that the simulation agents with a learned decision module can be generalized to various complex driving scenarios where the rule-based approach fails.
- 2
- 2
- 2
- GeoNorm: Unify Pre-Norm and Post-Norm with Geodesic Optimization1
GeoNorm is introduced, a novel method that replaces standard normalization with geodesic updates on the manifold that can be seamlessly integrated into standard Transformer architectures, achieving performance improvements with negligible additional computational cost.
- Diagnosing Shoulder Disorders Using Multimodal Large Language Models and Consumer-Grade Cameras1
This work focuses on the innovative application of Multimodal Large Language Models (MLLMs) in the preliminary diagnosis of shoulder disorders and proposes a Hybrid Motion Video Diagnosis framework (HMVDx), which divides the two tasks of action understanding and disease diagnosis, which are respectively completed by two MLLMs.
- NopeRoomGS: Indoor 3D Gaussian Splatting Optimization without Camera Pose Input1
This work introduces Nope-RoomGS, an optimization framework with no need for camera inputs, which effectively addresses the textureless regions and abrupt camera motion in indoor room environments through a local-to-global optimization paradigm for 3D GS reconstruction.
- 1
- –
- –
- –
- –
- –
- Pose-Agnostic Robotic Functional Grasping via Observation-Action Canonicalization–
AnyMug is proposed, a canonicalized visuomotor reinforcement learning framework for functional grasping that trains a single closed-loop policy entirely in simulation and deploys it zero-shot on a real robot.
- Wearable Intelligence for Healthcare Robotics: From Brain Activity to Body Movements–
This research aims to pioneer efficient and reliable wearable intelligence algorithms that transform healthcare robotics into adaptive, patient-centered systems and lays the groundwork for the next generation of healthcare technologies that are proactive, trustworthy, and deeply aligned with human well-being.
- Cubit: Token Mixer with Kernel Ridge Regression–
It is argued that Cubit, as a KRR-based architecture, provides a stronger mathematical foundation than the vanilla Transformer, whose attention mechanism corresponds to Nadaraya-Watson regression, and experimental results suggest that Cubit may exhibit stronger long-sequence modeling capability.
- SynapseRoute: an auto-route switching framework on dual-state large language model for medical question answering–
This study proposes SynapseRoute, an intelligent routing framework that dynamically assigns queries to either mode to optimise accuracy, latency and token cost and provides practical tools for deploying adaptive LLM systems with controllable performance–cost trade-offs.
- –
- –
- NodeOP: Optimizing Node Management for Decentralized Networks–
This work presents NodeOP, a novel framework designed to optimize the management of General Node Operators in decentralized networks by integrating Agent-Based Modeling with a Tendermint Byzantine Fault Tolerance-based consensus mechanism, which ensures stable equilibrium in node task distribution.
- –
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-15. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.