arXiv subjects

Dong Chen

Publications and source records attributed to Dong Chen.

At least 127 records · Page 7Linked to original sources

arXiv2024

Experimental Study of RCS Diversity with Novel No-divergent OAM Beams

This research proposes a novel approach utilizing Orbital Angular Momentum (OAM) beams to enhance Radar Cross Section (RCS) diversity for target detection in future transportation systems. Unlike conventional OAM beams with hollow-shaped divergence patterns, the new proposed OAM beams provide uniform illumination across the target without a central energy void, but keep the inherent phase gradient of vortex property. We utilize waveguide slot antennas to generate four different modes of these novel OAM beams at X-band frequency. Furthermore, these different mode OAM beams are used to illuminate metal models, and the resulting RCS is compared with that obtained using plane waves. The findings reveal that the novel OAM beams produce significant azimuthal RCS diversity, providing a new approach for the detection of weak and small targets.This study not only reveals the RCS diversity phenomenon based on novel OAM beams of different modes but also addresses the issue of energy divergence that hinders traditional OAM beams in long-range detection applications.

eess.SP↗

arXiv2024

CodeV: Issue Resolving with Visual Data

Large Language Models (LLMs) have advanced rapidly in recent years, with their applications in software engineering expanding to more complex repository-level tasks. GitHub issue resolving is a key challenge among these tasks. While recent approaches have made progress on this task, they focus on textual data within issues, neglecting visual data. However, this visual data is crucial for resolving issues as it conveys additional knowledge that text alone cannot. We propose CodeV, the first approach to leveraging visual data to enhance the issue-resolving capabilities of LLMs. CodeV resolves each issue by following a two-phase process: data processing and patch generation. To evaluate CodeV, we construct a benchmark for visual issue resolving, namely Visual SWE-bench. Through extensive experiments, we demonstrate the effectiveness of CodeV, as well as provide valuable insights into leveraging visual data to resolve GitHub issues.

cs.SE↗

arXiv2024

Category-Specific Topological Learning of Metal-Organic Frameworks

Metal-organic frameworks (MOFs) are porous, crystalline materials with high surface area, adjustable porosity, and structural tunability, making them ideal for diverse applications. However, traditional experimental and computational methods have limited scalability and interpretability, hindering effective exploration of MOF structure-property relationships. To address these challenges, we introduce, for the first time, a category-specific topological learning (CSTL), which combines algebraic topology with chemical insights for robust property prediction. The model represents MOF structures as simplicial complexes and incorporates elemental categorizations to enable balanced, interpretable machine learning study. By integrating category-specific persistent homology, CSTL captures both global and local structural characteristics, rendering multi-dimensional, category-specific descriptors that support a predictive model with high accuracy and robustness across eight MOF datasets, outperforming all previous results. This alignment of topological and chemical features enhances the predictive power and interpretability of CSTL, advancing understanding of structure-property relationships of MOFs and promoting efficient material discovery.

q-bio.BM↗

arXiv2024

Superionic Ionic Conductor Discovery via Multiscale Topological Learning

Lithium superionic conductors (LSICs) are crucial for next-generation solid-state batteries, offering exceptional ionic conductivity and enhanced safety for renewable energy and electric vehicles. However, their discovery is extremely challenging due to the vast chemical space, limited labeled data, and the understanding of complex structure-function relationships required for optimizing ion transport. This study introduces a multiscale topological learning (MTL) framework, integrating algebraic topology and unsupervised learning to tackle these challenges efficiently. By modeling lithium-only and lithium-free substructures, the framework extracts multiscale topological features and introduces two topological screening metrics-cycle density and minimum connectivity distance-to ensure structural connectivity and ion diffusion compatibility. Promising candidates are clustered via unsupervised algorithms to identify those resembling known superionic conductors. For final refinement, candidates that pass chemical screening undergo ab initio molecular dynamics simulations for validation. This approach led to the discovery of 14 novel LSICs, four of which have been independently validated in recent experiments. This success accelerates the identification of LSICs and demonstrates broad adaptability, offering a scalable tool for addressing complex materials discovery challenges.

cond-mat.mtrl-sci↗

arXiv2024

Evidence for reduced periodic lattice distortion within the Sb-terminated surface layer of the kagome metal CsV$_3$Sb$_5$

The discovery of the kagome metal CsV$_3$Sb$_5$ sparked broad interest, due to the coexistence of a charge density wave (CDW) phase and possible unconventional superconductivity in the material. In this study, we use low-energy electron diffraction (LEED) with a $μ$m-sized electron beam to explore the periodic lattice distortion at the antimony-terminated surface in the CDW phase. We recorded high-quality backscattering diffraction patterns in ultrahigh vacuum from multiple cleaved samples. Unexpectedly, we did not find superstructure reflexes at intensity levels predicted from dynamical LEED calculations for the reported $2 \times 2 \times 2$ bulk structure. Our results suggest that in CsV$_3$Sb$_5$ the periodic lattice distortion accompanying the CDW is less pronounced at Sb-terminated surfaces than in the bulk.

cond-mat.str-el↗

arXiv2024

CogACT: A Foundational Vision-Language-Action Model for Synergizing Cognition and Action in Robotic Manipulation

The advancement of large Vision-Language-Action (VLA) models has significantly improved robotic manipulation in terms of language-guided task execution and generalization to unseen scenarios. While existing VLAs adapted from pretrained large Vision-Language-Models (VLM) have demonstrated promising generalizability, their task performance is still unsatisfactory as indicated by the low tasks success rates in different environments. In this paper, we present a new advanced VLA architecture derived from VLM. Unlike previous works that directly repurpose VLM for action prediction by simple action quantization, we propose a omponentized VLA architecture that has a specialized action module conditioned on VLM output. We systematically study the design of the action module and demonstrates the strong performance enhancement with diffusion action transformers for action sequence modeling, as well as their favorable scaling behaviors. We also conduct comprehensive experiments and ablation studies to evaluate the efficacy of our models with varied designs. The evaluation on 5 robot embodiments in simulation and real work shows that our model not only significantly surpasses existing VLAs in task performance and but also exhibits remarkable adaptation to new robots and generalization to unseen objects and backgrounds. It exceeds the average success rates of OpenVLA which has similar model size (7B) with ours by over 35% in simulated evaluation and 55% in real robot experiments. It also outperforms the large RT-2-X model (55B) by 18% absolute success rates in simulation. Code and models can be found on our project page (https://cogact.github.io/).

cs.RO↗

arXiv2024

The Critical Canvas--How to regain information autonomy in the AI era

In the era of AI, recommendation algorithms and generative AI challenge information autonomy by creating echo chambers and blurring the line between authentic and fabricated content. The Critical Canvas addresses these challenges with a novel information exploration platform designed to restore balance between algorithmic efficiency and human agency. It employs three key mechanisms: multi-dimensional exploration across logical, temporal, and geographical perspectives; dynamic knowledge entry generation to capture complex relationships between concepts; and a phase space to evaluate the credibility of both the content and its sources. Particularly relevant to technical AI governance, where stakeholders must navigate intricate specifications and safety frameworks, the platform transforms overwhelming technical information into actionable insights. The Critical Canvas empowers users to regain autonomy over their information consumption through structured yet flexible exploration pathways, creative visualization, human-centric navigation, and transparent source evaluation. It fosters a comprehensive understanding of nuanced topics, enabling more informed decision-making and effective policy development in the age of AI.

cs.CY↗

arXiv2024

Robotic transcatheter tricuspid valve replacement with hybrid enhanced intelligence: a new paradigm and first-in-vivo study

Transcatheter tricuspid valve replacement (TTVR) is the latest treatment for tricuspid regurgitation and is in the early stages of clinical adoption. Intelligent robotic approaches are expected to overcome the challenges of surgical manipulation and widespread dissemination, but systems and protocols with high clinical utility have not yet been reported. In this study, we propose a complete solution that includes a passive stabilizer, robotic drive, detachable delivery catheter and valve manipulation mechanism. Working towards autonomy, a hybrid augmented intelligence approach based on reinforcement learning, Monte Carlo probabilistic maps and human-robot co-piloted control was introduced. Systematic tests in phantom and first-in-vivo animal experiments were performed to verify that the system design met the clinical requirement. Furthermore, the experimental results confirmed the advantages of co-piloted control over conventional master-slave control in terms of time efficiency, control efficiency, autonomy and stability of operation. In conclusion, this study provides a comprehensive pathway for robotic TTVR and, to our knowledge, completes the first animal study that not only successfully demonstrates the application of hybrid enhanced intelligence in interventional robotics, but also provides a solution with high application value for a cutting-edge procedure.

cs.RO↗

arXiv2024

A Retrospective on the Robot Air Hockey Challenge: Benchmarking Robust, Reliable, and Safe Learning Techniques for Real-world Robotics

Machine learning methods have a groundbreaking impact in many application domains, but their application on real robotic platforms is still limited. Despite the many challenges associated with combining machine learning technology with robotics, robot learning remains one of the most promising directions for enhancing the capabilities of robots. When deploying learning-based approaches on real robots, extra effort is required to address the challenges posed by various real-world factors. To investigate the key factors influencing real-world deployment and to encourage original solutions from different researchers, we organized the Robot Air Hockey Challenge at the NeurIPS 2023 conference. We selected the air hockey task as a benchmark, encompassing low-level robotics problems and high-level tactics. Different from other machine learning-centric benchmarks, participants need to tackle practical challenges in robotics, such as the sim-to-real gap, low-level control issues, safety problems, real-time requirements, and the limited availability of real-world data. Furthermore, we focus on a dynamic environment, removing the typical assumption of quasi-static motions of other real-world benchmarks. The competition's results show that solutions combining learning-based approaches with prior knowledge outperform those relying solely on data when real-world deployment is challenging. Our ablation study reveals which real-world factors may be overlooked when building a learning-based solution. The successful real-world air hockey deployment of best-performing agents sets the foundation for future competitions and follow-up research directions.

cs.RO↗

arXiv2024

A Comparative Study of Deep Reinforcement Learning for Crop Production Management

Crop production management is essential for optimizing yield and minimizing a field's environmental impact to crop fields, yet it remains challenging due to the complex and stochastic processes involved. Recently, researchers have turned to machine learning to address these complexities. Specifically, reinforcement learning (RL), a cutting-edge approach designed to learn optimal decision-making strategies through trial and error in dynamic environments, has emerged as a promising tool for developing adaptive crop management policies. RL models aim to optimize long-term rewards by continuously interacting with the environment, making them well-suited for tackling the uncertainties and variability inherent in crop management. Studies have shown that RL can generate crop management policies that compete with, and even outperform, expert-designed policies within simulation-based crop models. In the gym-DSSAT crop model environment, one of the most widely used simulators for crop management, proximal policy optimization (PPO) and deep Q-networks (DQN) have shown promising results. However, these methods have not yet been systematically evaluated under identical conditions. In this study, we evaluated PPO and DQN against static baseline policies across three different RL tasks, fertilization, irrigation, and mixed management, provided by the gym-DSSAT environment. To ensure a fair comparison, we used consistent default parameters, identical reward functions, and the same environment settings. Our results indicate that PPO outperforms DQN in fertilization and irrigation tasks, while DQN excels in the mixed management task. This comparative analysis provides critical insights into the strengths and limitations of each approach, advancing the development of more effective RL-based crop management strategies.

eess.SY↗

arXiv2024

GaussianCube: A Structured and Explicit Radiance Representation for 3D Generative Modeling

We introduce a radiance representation that is both structured and fully explicit and thus greatly facilitates 3D generative modeling. Existing radiance representations either require an implicit feature decoder, which significantly degrades the modeling power of the representation, or are spatially unstructured, making them difficult to integrate with mainstream 3D diffusion methods. We derive GaussianCube by first using a novel densification-constrained Gaussian fitting algorithm, which yields high-accuracy fitting using a fixed number of free Gaussians, and then rearranging these Gaussians into a predefined voxel grid via Optimal Transport. Since GaussianCube is a structured grid representation, it allows us to use standard 3D U-Net as our backbone in diffusion modeling without elaborate designs. More importantly, the high-accuracy fitting of the Gaussians allows us to achieve a high-quality representation with orders of magnitude fewer parameters than previous structured representations for comparable quality, ranging from one to two orders of magnitude. The compactness of GaussianCube greatly eases the difficulty of 3D generative modeling. Extensive experiments conducted on unconditional and class-conditioned object generation, digital avatar creation, and text-to-3D synthesis all show that our model achieves state-of-the-art generation results both qualitatively and quantitatively, underscoring the potential of GaussianCube as a highly accurate and versatile radiance representation for 3D generative modeling. Project page: https://gaussiancube.github.io/.

cs.CV↗

arXiv2024

Simplified Diffusion Schrödinger Bridge

This paper introduces a novel theoretical simplification of the Diffusion Schrödinger Bridge (DSB) that facilitates its unification with Score-based Generative Models (SGMs), addressing the limitations of DSB in complex data generation and enabling faster convergence and enhanced performance. By employing SGMs as an initial solution for DSB, our approach capitalizes on the strengths of both frameworks, ensuring a more efficient training process and improving the performance of SGM. We also propose a reparameterization technique that, despite theoretical approximations, practically improves the network's fitting capabilities. Our extensive experimental evaluations confirm the effectiveness of the simplified DSB, demonstrating its significant improvements. We believe the contributions of this work pave the way for advanced generative modeling.

cs.LG↗

arXiv2024

Multi-Agent Reinforcement Learning for Connected and Automated Vehicles Control: Recent Advancements and Future Prospects

Connected and automated vehicles (CAVs) are considered a potential solution for future transportation challenges, aiming to develop systems that are efficient, safe, and environmentally friendly. However, CAV control presents significant challenges due to the complexity of interconnectivity and coordination required among vehicles. Multi-agent reinforcement learning (MARL), which has shown notable advancements in addressing complex problems in autonomous driving, robotics, and human-vehicle interaction, emerges as a promising tool to enhance CAV capabilities. Despite its potential, there is a notable absence of current reviews on mainstream MARL algorithms for CAVs. To fill this gap, this paper offers a comprehensive review of MARL's application in CAV control. The paper begins with an introduction to MARL, explaining its unique advantages in handling complex and multi-agent scenarios. It then presents a detailed survey of MARL applications across various control dimensions for CAVs, including critical scenarios such as platooning control, lane-changing, and unsignalized intersections. Additionally, the paper reviews prominent simulation platforms essential for developing and testing MARL algorithms. Lastly, it examines the current challenges in deploying MARL for CAV control, including macro-micro optimization, communication, mixed traffic, and sim-to-real challenges. Potential solutions discussed include hierarchical MARL, decentralized MARL, adaptive interactions, and offline MARL.

cs.RO↗

arXiv2024

Integrating Reinforcement Learning and Large Language Models for Crop Production Process Management Optimization and Control through A New Knowledge-Based Deep Learning Paradigm

Efficient and sustainable crop production process management is crucial to meet the growing global demand for food, fuel, and feed while minimizing environmental impacts. Traditional crop management practices, often developed through empirical experience, face significant challenges in adapting to the dynamic nature of modern agriculture, which is influenced by factors such as climate change, soil variability, and market conditions. Recently, reinforcement learning (RL) and large language models (LLMs) bring transformative potential, with RL providing adaptive methodologies to learn optimal strategies and LLMs offering vast, superhuman knowledge across agricultural domains, enabling informed, context-specific decision-making. This paper systematically examines how the integration of RL and LLMs into crop management decision support systems (DSSs) can drive advancements in agricultural practice. We explore recent advancements in RL and LLM algorithms, their application within crop management, and the use of crop management simulators to develop these technologies. The convergence of RL and LLMs with crop management DSSs presents new opportunities to optimize agricultural practices through data-driven, adaptive solutions that can address the uncertainties and complexities of crop production. However, this integration also brings challenges, particularly in real-world deployment. We discuss these challenges and propose potential solutions, including the use of offline RL and enhanced LLM integration, to maximize the effectiveness and sustainability of crop management. Our findings emphasize the need for continued research and innovation to unlock the full potential of these advanced tools in transforming agricultural systems into optimal and controllable ones.

eess.SY↗

arXiv2024

SynChart: Synthesizing Charts from Language Models

With the release of GPT-4V(O), its use in generating pseudo labels for multi-modality tasks has gained significant popularity. However, it is still a secret how to build such advanced models from its base large language models (LLMs). This work explores the potential of using LLMs alone for data generation and develop competitive multi-modality models focusing on chart understanding. We construct a large-scale chart dataset, SynChart, which contains approximately 4 million diverse chart images with over 75 million dense annotations, including data tables, code, descriptions, and question-answer sets. We trained a 4.2B chart-expert model using this dataset and achieve near-GPT-4O performance on the ChartQA task, surpassing GPT-4V.

cs.AI↗

arXiv2024

Back-stepping Experience Replay with Application to Model-free Reinforcement Learning for a Soft Snake Robot

In this paper, we propose a novel technique, Back-stepping Experience Replay (BER), that is compatible with arbitrary off-policy reinforcement learning (RL) algorithms. BER aims to enhance learning efficiency in systems with approximate reversibility, reducing the need for complex reward shaping. The method constructs reversed trajectories using back-stepping transitions to reach random or fixed targets. Interpretable as a bi-directional approach, BER addresses inaccuracies in back-stepping transitions through a distillation of the replay experience during learning. Given the intricate nature of soft robots and their complex interactions with environments, we present an application of BER in a model-free RL approach for the locomotion and navigation of a soft snake robot, which is capable of serpentine motion enabled by anisotropic friction between the body and ground. In addition, a dynamic simulator is developed to assess the effectiveness and efficiency of the BER algorithm, in which the robot demonstrates successful learning (reaching a 100% success rate) and adeptly reaches random targets, achieving an average speed 48% faster than that of the best baseline approach.

cs.RO↗

arXiv2024

Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone

We introduce phi-3-mini, a 3.8 billion parameter language model trained on 3.3 trillion tokens, whose overall performance, as measured by both academic benchmarks and internal testing, rivals that of models such as Mixtral 8x7B and GPT-3.5 (e.g., phi-3-mini achieves 69% on MMLU and 8.38 on MT-bench), despite being small enough to be deployed on a phone. Our training dataset is a scaled-up version of the one used for phi-2, composed of heavily filtered publicly available web data and synthetic data. The model is also further aligned for robustness, safety, and chat format. We also provide parameter-scaling results with a 7B, 14B models trained for 4.8T tokens, called phi-3-small, phi-3-medium, both significantly more capable than phi-3-mini (e.g., respectively 75%, 78% on MMLU, and 8.7, 8.9 on MT-bench). To enhance multilingual, multimodal, and long-context capabilities, we introduce three models in the phi-3.5 series: phi-3.5-mini, phi-3.5-MoE, and phi-3.5-Vision. The phi-3.5-MoE, a 16 x 3.8B MoE model with 6.6 billion active parameters, achieves superior performance in language reasoning, math, and code tasks compared to other open-source models of similar scale, such as Llama 3.1 and the Mixtral series, and on par with Gemini-1.5-Flash and GPT-4o-mini. Meanwhile, phi-3.5-Vision, a 4.2 billion parameter model derived from phi-3.5-mini, excels in reasoning tasks and is adept at handling both single-image and text prompts, as well as multi-image and text prompts.

cs.CL↗

arXiv2024

SWE-bench-java: A GitHub Issue Resolving Benchmark for Java

GitHub issue resolving is a critical task in software engineering, recently gaining significant attention in both industry and academia. Within this task, SWE-bench has been released to evaluate issue resolving capabilities of large language models (LLMs), but has so far only focused on Python version. However, supporting more programming languages is also important, as there is a strong demand in industry. As a first step toward multilingual support, we have developed a Java version of SWE-bench, called SWE-bench-java. We have publicly released the dataset, along with the corresponding Docker-based evaluation environment and leaderboard, which will be continuously maintained and updated in the coming months. To verify the reliability of SWE-bench-java, we implement a classic method SWE-agent and test several powerful LLMs on it. As is well known, developing a high-quality multi-lingual benchmark is time-consuming and labor-intensive, so we welcome contributions through pull requests or collaboration to accelerate its iteration and refinement, paving the way for fully automated programming.

cs.SE↗