Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks35 from public data
- Towards Human-Level Bimanual Dexterous Manipulation with Reinforcement Learning217
The results show that the PPO type of on-policy algorithms can master simple manipulation tasks that are equivalent up to 48-month human babies, while multi-agent RL can further help to master manipulations that require skilled bimanual cooperation.
- D$^2$NeRF: Self-Supervised Decoupling of Dynamic and Static Objects from a Monocular Video201
This work introduces Decoupled Dynamic Neural Radiance Field (D$^2$NeRF), a self-supervised approach that takes a monocular video and learns a 3D scene representation which decouples moving objects, including their shadows, from the static background.
- Influence of physicochemical properties and starch fine structure on the eating quality of hybrid rice with similar apparent amylose content114
The physicochemical properties and starch structures of hybrid rice varieties with similar apparent amylose content but different taste values are compared to indicate that a greater diversity of forms and sizes of starch granules might influence the eating quality of hybrid Rice.
- Domain Randomization-Enhanced Depth Simulation and Restoration for Perceiving and Grasping Specular and Transparent Objects95
This work proposes a powerful RGBD fusion network, SwinDRNet, and proposes Domain Randomization-Enhanced Depth Simulation (DREDS) approach to simulate an active stereo depth system using physically based rendering and generates a large-scale synthetic dataset that contains 130K photorealistic RGB images along with their simulated depths carrying realistic sensor noises.
- 87
- Pairwise Proximal Policy Optimization: Harnessing Relative Feedback for LLM Alignment83
A new framework, reinforcement learning with relative feedback, and a novel trajectory-wise policy gradient algorithm, Pairwise Proximal Policy Optimization (P3O) that operates directly on comparative rewards are proposed that can align with human preferences as well as or better than prior methods.
- Structure and in vitro digestibility on complex of corn starch with soy isoflavone64
The digestibility and pGI of starch can be improved by complexed with soy isoflavone, as indicated by in vitro digestion model.
- 32
- Computing Approximate Graph Edit Distance via Optimal Transport15
An ensemble approach that integrates a supervised learning-based method and an unsupervised method, both based on optimal transport, that significantly outperform the existing methods in terms of the performance of GED computation, edit path generation, and model generalizability is proposed.
- PyPose v0.6: The Imperative Programming Interface for Robotics12
This work presents the fundamental design principle of the imperative programming interface, and showcases the flexible usage of diverse functionalities and modules in PyPose using an extremely simple Dubins car example.
- 10
- Neural Fields with Hard Constraints of Arbitrary Differential Order10
A series of approaches for enforcing hard constraints on neural fields, which are referred to as Constrained Neural Fields (CNF), and a framework that enables highly efficient model and constraint specification, which can be readily applied to any downstream task where hard constraints need to be explicitly satisfied during optimization.
- Correlations Between Parental Lines and Indica Hybrid Rice in Terms of Eating Quality Traits10
This study provides a framework for the development of hybrid rice with superior eating quality by analyzing the relationship between hybrid combinations and parental lines and found that the eating quality traits of hybrid combinations were determined by both parents.
- ARF-Plus: Controlling Perceptual Factors in Artistic Radiance Fields for 3D Scene Stylization7
ARF-Plus is presented, a unique 3D neural style transfer framework offering manageable control over perceptual factors, to systematically explore the perceptual controllability in 3D scene stylization.
- Boosting Universal LLM Reward Design through Heuristic Reward Observation Space Evolution6
A novel heuristic framework is proposed that enhances LLM-driven reward design by evolving the ROS through a table-based exploration caching mechanism and a text-code reconciliation strategy, overcoming the Markovian constraint typically found in LLM dialogues and facilitating more effective exploration.
- Gaussian Head & Shoulders: High Fidelity Neural Upper Body Avatars with Anchor Gaussian Guided Texture Warping6
It is demonstrated that Gaussian Head&Shoulders can fit the high-frequency details on the clothed upper body with high fidelity and potentially improve the accuracy and fidelity of the head region and is evaluated with casual phone-captured and internet videos.
- 6
- Corrigendum: Correlations Between Parental Lines and Indica Hybrid Rice in Terms of Eating Quality Traits6
This research presents a novel probabilistic approach to estimating the number of mitochondria in the response of the immune system to certain types of infectious disease.
- 3
- LTOS: Layout-controllable Text-object Synthesis via Adaptive Cross-attention Fusions2
This paper introduces a unified task called layout-controllable text-object synthesis (LTOS), which merges text rendering and layout-based object generation, and proposes a self-adaptive cross-attention fusion that forces the image generation to attend more to important text spans.
- 2
- Feasibility Restoration under Conflicting STL Specifications with Pareto-Optimal Refinement1
This paper proposes a unified two-stage framework that first restores feasibility via minimal relaxation, then refine the feasible solution by formulating it as a value-aware multi-objective optimization problem, approximate the Pareto front of the multi-objective optimization.
- 1
- Enhancing the Performance of DeepReach on High-Dimensional Systems through Optimizing Activation Functions1
This paper aims to improve the performance of DeepReach on high-dimensional systems by exploring different choices of activation functions, and demonstrates the effectiveness of the approach on a 9D multi-vehicle collision problem.
- ClipSim: A GPU-friendly Parallel Framework for Single-Source SimRank with Accuracy Guarantee1
This paper proposes ClipSim, the first GPU-friendly parallel framework that accelerates the single-source SimRank on GPU with accuracy guarantee, and designs a novel data structure and GPU- friendly parallel algorithms for efficient computation of all the operations of Sim Rank on GPU.
- NodeBits: A Plug-and-Play Framework for Accelerating Graph Inference by Post-Hoc Binary Quantization–
This paper proposes NodeBits, a lightweight plug-and-play framework that accelerates graph inference via a novel post-hoc binary quantization scheme that consistently outperforms existing GNN-to-MLP methods in both accuracy and inference speed.
- CompreGel: Efficient Distributed Graph Propagation via Error-Bounded Lossy Message Compression–
This work proposes a message compression framework for Pregel-like systems, called CompreGel, which accelerates graph propagation algorithms by significantly reducing the communication cost of message passing and theoretically analyzes how to set the error bound parameter of SZ properly to bound the accumulative error of vertex values across iterations of graph propagation.
- –
- –
- –
- Efficient Enumeration of Large Maximal k-Plexes–
This paper proposes an efficient branch-and-bound algorithm as well as its task-based parallel version to enumerate all maximal k -plexes with at least q vertices, and adopts an effective search space partitioning approach that provides a good time complexity.
- –
- –
- –
- –
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 2026-10-11. 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.