arXiv subjects

Rui Chen

Publications and source records attributed to Rui Chen.

At least 181 records · Page 10Linked to original sources

arXiv2025

LLMControl: Grounded Control of Text-to-Image Diffusion-based Synthesis with Multimodal LLMs

Recent spatial control methods for text-to-image (T2I) diffusion models have shown compelling results. However, these methods still fail to precisely follow the control conditions and generate the corresponding images, especially when encountering the textual prompts that contain multiple objects or have complex spatial compositions. In this work, we present a LLM-guided framework called LLM\_Control to address the challenges of the controllable T2I generation task. By improving grounding capabilities, LLM\_Control is introduced to accurately modulate the pre-trained diffusion models, where visual conditions and textual prompts influence the structures and appearance generation in a complementary way. We utilize the multimodal LLM as a global controller to arrange spatial layouts, augment semantic descriptions and bind object attributes. The obtained control signals are injected into the denoising network to refocus and enhance attention maps according to novel sampling constraints. Extensive qualitative and quantitative experiments have demonstrated that LLM\_Control achieves competitive synthesis quality compared to other state-of-the-art methods across various pre-trained T2I models. It is noteworthy that LLM\_Control allows the challenging input conditions on which most of the existing methods

cs.CV↗

arXiv2025

BadVideo: Stealthy Backdoor Attack against Text-to-Video Generation

Text-to-video (T2V) generative models have rapidly advanced and found widespread applications across fields like entertainment, education, and marketing. However, the adversarial vulnerabilities of these models remain rarely explored. We observe that in T2V generation tasks, the generated videos often contain substantial redundant information not explicitly specified in the text prompts, such as environmental elements, secondary objects, and additional details, providing opportunities for malicious attackers to embed hidden harmful content. Exploiting this inherent redundancy, we introduce BadVideo, the first backdoor attack framework tailored for T2V generation. Our attack focuses on designing target adversarial outputs through two key strategies: (1) Spatio-Temporal Composition, which combines different spatiotemporal features to encode malicious information; (2) Dynamic Element Transformation, which introduces transformations in redundant elements over time to convey malicious information. Based on these strategies, the attacker's malicious target seamlessly integrates with the user's textual instructions, providing high stealthiness. Moreover, by exploiting the temporal dimension of videos, our attack successfully evades traditional content moderation systems that primarily analyze spatial information within individual frames. Extensive experiments demonstrate that BadVideo achieves high attack success rates while preserving original semantics and maintaining excellent performance on clean inputs. Overall, our work reveals the adversarial vulnerability of T2V models, calling attention to potential risks and misuse. Our project page is at https://wrt2000.github.io/BadVideo2025/.

cs.CV↗

arXiv2025

Emergence of charm-strange dibaryons with negative parity via baryon-baryon interactions

Within the framework of the one-boson-exchange model, we systematically perform a coupled channel analysis of the $P-$wave interactions between a charm baryon and light baryon, the involved channels include $Ξ_cN$, $Λ_cΣ$, $Ξ_c^{\prime}N$, $Σ_cΛ$, $Ξ_c^*N$, $Σ_c^*Λ$, $Σ_cΣ$, and $Σ_c^*Σ$. Our results can predict several possible molecular candidates, such as a $Ξ_c^{\prime}N$ molecule with $I(J^P)=0(1^-)$, $Ξ_c^{*}N$ molecules with $0(0^-, 1^-, 2^-)$, $Σ_cΣ$ molecules with $0(1^-)$ and $1(1^-)$, and $Σ_c^*Σ$ molecules with $0(0^-, 1^-, 2^-)$, and $1(2^-)$. The coupled channel effects significantly influence the formation of the $Σ_cΣ$ molecule with $1(1^-)$. Furthermore, we analyze the phase shifts for these coupled channel systems. Our analysis not only confirms the existence of the predicted molecules but also identifies potential resonant dibaryons, including a $Σ_cΣ$ shape-type resonance with $1(1^-)$, a $Σ_c^*Σ$ shape-type resonance with $1(0^-)$, a $Λ_cΣ/Σ_cΣ$ coupled Feshbach-type resonance with $1(1^-)$, and $Λ_cΣ/Σ_c^*Σ$ coupled Feshbach-type resonances with $1(0^-, 2^-)$.

hep-ph↗

arXiv2025

Observation of a Knotted Electron Diffusion Region in Earth's Magnetotail Reconnection

Magnetic reconnection is a fundamental plasma process that alters the magnetic field topology and releases magnetic energy. Most numerical simulations and spacecraft observations assume a two-dimensional diffusion region, with the electron diffusion region (EDR) embedded in the same plane as the ion diffusion region (IDR) and a uniform guide field throughout. Using observations from Magnetospheric Multiscale (MMS) mission, we report a non-coplanar, knotted EDR in Earth's magnetotail current sheet. The reconnection plane of the knotted EDR deviates by approximately 38° from that of the IDR, with the guide field exhibiting both a 38° directional shift and a twofold increase in amplitude. Moreover, the Hall magnetic field is bipolar in the EDR but quadrupolar in the IDR, indicating different Hall current structures at electron and ion scales. These observations highlight the importance of three-dimensional effects and illustrate the complexity of multiscale coupling between the EDR and IDR during reconnection studies.1

physics.space-ph↗

arXiv2025

DexSim2Real$^{2}$: Building Explicit World Model for Precise Articulated Object Dexterous Manipulation

Articulated objects are ubiquitous in daily life. In this paper, we present DexSim2Real$^{2}$, a novel framework for goal-conditioned articulated object manipulation. The core of our framework is constructing an explicit world model of unseen articulated objects through active interactions, which enables sampling-based model predictive control to plan trajectories achieving different goals without requiring demonstrations or RL. It first predicts an interaction using an affordance network trained on self-supervised interaction data or videos of human manipulation. After executing the interactions on the real robot to move the object parts, we propose a novel modeling pipeline based on 3D AIGC to build a digital twin of the object in simulation from multiple frames of observations. For dexterous hands, we utilize eigengrasp to reduce the action dimension, enabling more efficient trajectory searching. Experiments validate the framework's effectiveness for precise manipulation using a suction gripper, a two-finger gripper and two dexterous hand. The generalizability of the explicit world model also enables advanced manipulation strategies like manipulating with tools.

cs.RO↗

arXiv2025

Godement--Jacquet L-function and homological theta lifting

In this paper we investigate the theta lifting of type II dual pairs over a non-Archimedean local field, by combining the homological method of Adams--Prasad--Savin and the analytic method of Fang--Sun--Xue. We have three main results: 1. we determine completely the big theta lift of an irreducible representation when its Godement--Jacquet L-function is holomorphic at a critical point; 2. we compute the big theta lift of all characters, hence determine the space of eigendistributions on matrix spaces for all characters; 3. we show that the Weil representation is projective if and only if the dual pair is almost in the stable range.

math.RT↗

arXiv2025

Non-tempered Gan-Gross-Prasad conjecture for Archimedean general linear groups

The local non-tempered Gan-Gross-Prasad conjecture suggests that, for a pair of irreducible Arthur type representations of two successive general linear groups, they have a non-zero Rankin-Selberg period if and only if they are "relevant". In this paper, we prove the "period implies relevance" direction of this conjecture for general linear groups over Archimedean local fields. Our proof is based on a previous result of Gourevitch-Sayag-Karshon on the annihilator varieties of distinguished representations. Combining with a recent result of P. Boisseau on the direction "relevance implies period'', this conjecture for general linear groups over Archimedean local fields is settled.

math.RT↗

arXiv2025

Exploring the two-body strong decay properties of the possible $Λ_cK^{}$ and $Σ_cK^{()}$ molecules

In this work, we apply the effective Lagrangian approach to investigate the two-body strong decay behaviors of the possible $Λ_c K^*$ and $Σ_c K^{(*)}$ molecules, as predicted in our previous study [\href{https://doi.org/10.1103/PhysRevD.108.054011}{Phys. Rev. D \textbf{108}, 054011 (2023)}]. Our results indicate that the decay width for the coupled $Σ_c K / Λ_c K^* / Σ_c K^*$ molecule with $I(J^P) = 1/2(1/2^-)$ is on the order of several MeV, with the $D_s N$ channel being dominant. For the coupled $Λ_c K^* / Σ_c K^*$ molecule with $1/2(1/2^-, 3/2^-)$, the decay widths are on the order of tens of MeV, with the dominant channels being $Σ_c K$ and $Σ_c^* K$, respectively. For the $Σ_c K^*$ molecules with $1/2(1/2^-)$, the decay width can reach one hundred MeV, with $Σ_c K$ and $Λ_c K$ being the dominant decay channels. The decay widths for the $Σ_c K^*$ molecules with $1/2(3/2^-)$ and $3/2(1/2^-)$ are on the order of tens of MeV, with the dominant decay modes being $Σ_c^* K$ and $Σ_c K$, respectively. The branching ratios for all the discussed channels show little dependence on the binding energies.

hep-ph↗

arXiv2025

NEO-PGA: Nonvolatile electro-optically programmable gate array

Programmable photonic integrated circuits (PICs) offer a unique opportunity to create a flexible platform, akin to electronic field programmable gate array (FPGA). These photonic PGAs can implement versatile functionalities for applications ranging from optical interconnects to microwave photonics. However, state-of-the-art programmable photonics relies predominantly on volatile thermo-optic tuning, which suffers from high static power consumption, large footprints, and thermal crosstalk. All these dramatically limit the gate density and pose a fundamental limit to the scalability. Chalcogenide-based phase-change materials (PCMs) offer a superior alternative due to their nonvolatility and substantial optical contrast, though challenges such as optical loss, and bit precision severely limited their application in large-scale PICs. Here, we demonstrate precise, multi-bit, low-loss tuning of the emerging PCM Sb2Se3 using a closed-loop, "program-and-verify" method. Electrically reconfigurable PCM-integrated silicon photonic gates are implemented on a 300mm silicon photonic platform, using circulating and forward Mach-Zehnder interferometer (MZI) meshes. In the circulating mesh, we realize broadband optical switching fabrics and high-Q coupled resonators with unprecedented local control of coupling rates, which further enable exploration of coupled-cavity systems. The forward mesh supports self-configurable MZIs that sort two orthogonal beams to different ports. These results showcase a new type of scalable photonic PGA enabled by PCMs, offering a pathway toward general-purpose, on-chip programmable photonic systems.

physics.optics↗

arXiv2025

Near-visible low power tuning of nematic-liquid crystal integrated silicon nitride ring resonator

The development of compact, low-power, and high-performance integrated photonic phase shifters is critical for advancing emerging technologies such as light detection and ranging (LiDAR), optical information processing and quantum applications. Liquid crystal (LC)-based phase shifters offer a promising solution thanks to their large refractive index contrast and their low power consumption. However, it remains challenging to incorporate LCs into integrated photonics and the operating wavelength has been limited to near infrared. Here, we demonstrate a liquid-crystal-based phase shifter operating at 780 nm, a relevant wavelength for cold atom-based quantum applications, by incorporating nematic LCs (E7) into a silicon nitride (SiN) ring resonator. Our device achieves 2pi phase modulation with very low power of 2.1 nW and low driving voltages of 7 V with exceptionally low Vpi times L (half wave voltage times length) value of 0.014 V-cm, enabling precise control over light propagation in a compact footprint. This work marks a significant step toward realizing low-power, compact, and efficient LC integrated photonic circuits based on SiN platform for next-generation LiDAR and quantum optical systems.

physics.optics↗

arXiv2025

Electrically reconfigurable nonvolatile flatband absorbers in the mid-infrared with wide spectral tuning range

While recent advances in reconfigurable photonics have provided new avenues for manipulating light on the subwavelength scale, on-demand control of infrared absorption remains elusive. Here, we experimentally demonstrate a plasmonic metasurface based on the phase change material Ge2Sb2Te5 with in-situ electrically-switchable infrared absorption in the 3-5 microns wavelength range. Unlike traditional infrared microstructures based on volatile phase change materials, our device does not require the external stimuli to be continuously applied in order to maintain a given optical state, thus enabling zero static power operation. Furthermore, the 400x deep-subwavelength field localization supported by our device not only allows robust tuning of its spectral response but also makes its absorptivity independent of the angle of incidence, thus enabling a flatband behavior. We conduct switching of our device using rapid thermal annealing and reversible switching using electrical pulses over 26 cycles. Our device provides new avenues for infrared absorption control and serves as a steppingstone for the next generation of mid-wave infrared photonics.

physics.optics↗

arXiv2025

Learn With Imagination: Safe Set Guided State-wise Constrained Policy Optimization

Deep reinforcement learning (RL) excels in various control tasks, yet the absence of safety guarantees hampers its real-world applicability. In particular, explorations during learning usually results in safety violations, while the RL agent learns from those mistakes. On the other hand, safe control techniques ensure persistent safety satisfaction but demand strong priors on system dynamics, which is usually hard to obtain in practice. To address these problems, we present Safe Set Guided State-wise Constrained Policy Optimization (S-3PO), a pioneering algorithm generating state-wise safe optimal policies with zero training violations, i.e., learning without mistakes. S-3PO first employs a safety-oriented monitor with black-box dynamics to ensure safe exploration. It then enforces an "imaginary" cost for the RL agent to converge to optimal behaviors within safety constraints. S-3PO outperforms existing methods in high-dimensional robotics tasks, managing state-wise constraints with zero training violation. This innovation marks a significant stride towards real-world safe RL deployment.

cs.RO↗

arXiv2025

Continual Learning and Lifting of Koopman Dynamics for Linear Control of Legged Robots

The control of legged robots, particularly humanoid and quadruped robots, presents significant challenges due to their high-dimensional and nonlinear dynamics. While linear systems can be effectively controlled using methods like Model Predictive Control (MPC), the control of nonlinear systems remains complex. One promising solution is the Koopman Operator, which approximates nonlinear dynamics with a linear model, enabling the use of proven linear control techniques. However, achieving accurate linearization through data-driven methods is difficult due to issues like approximation error, domain shifts, and the limitations of fixed linear state-space representations. These challenges restrict the scalability of Koopman-based approaches. This paper addresses these challenges by proposing a continual learning algorithm designed to iteratively refine Koopman dynamics for high-dimensional legged robots. The key idea is to progressively expand the dataset and latent space dimension, enabling the learned Koopman dynamics to converge towards accurate approximations of the true system dynamics. Theoretical analysis shows that the linear approximation error of our method converges monotonically. Experimental results demonstrate that our method achieves high control performance on robots like Unitree G1/H1/A1/Go2 and ANYmal D, across various terrains using simple linear MPC controllers. This work is the first to successfully apply linearized Koopman dynamics for locomotion control of high-dimensional legged robots, enabling a scalable model-based control solution.

cs.RO↗

arXiv2025

CraftsMan3D: High-fidelity Mesh Generation with 3D Native Generation and Interactive Geometry Refiner

We present a novel generative 3D modeling system, coined CraftsMan, which can generate high-fidelity 3D geometries with highly varied shapes, regular mesh topologies, and detailed surfaces, and, notably, allows for refining the geometry in an interactive manner. Despite the significant advancements in 3D generation, existing methods still struggle with lengthy optimization processes, irregular mesh topologies, noisy surfaces, and difficulties in accommodating user edits, consequently impeding their widespread adoption and implementation in 3D modeling software. Our work is inspired by the craftsman, who usually roughs out the holistic figure of the work first and elaborates the surface details subsequently. Specifically, we employ a 3D native diffusion model, which operates on latent space learned from latent set-based 3D representations, to generate coarse geometries with regular mesh topology in seconds. In particular, this process takes as input a text prompt or a reference image and leverages a powerful multi-view (MV) diffusion model to generate multiple views of the coarse geometry, which are fed into our MV-conditioned 3D diffusion model for generating the 3D geometry, significantly improving robustness and generalizability. Following that, a normal-based geometry refiner is used to significantly enhance the surface details. This refinement can be performed automatically, or interactively with user-supplied edits. Extensive experiments demonstrate that our method achieves high efficacy in producing superior-quality 3D assets compared to existing methods. HomePage: https://craftsman3d.github.io/, Code: https://github.com/wyysf-98/CraftsMan

cs.GR↗

arXiv2025

ManiSkill3: GPU Parallelized Robotics Simulation and Rendering for Generalizable Embodied AI

Simulation has enabled unprecedented compute-scalable approaches to robot learning. However, many existing simulation frameworks typically support a narrow range of scenes/tasks and lack features critical for scaling generalizable robotics and sim2real. We introduce and open source ManiSkill3, the fastest state-visual GPU parallelized robotics simulator with contact-rich physics targeting generalizable manipulation. ManiSkill3 supports GPU parallelization of many aspects including simulation+rendering, heterogeneous simulation, pointclouds/voxels visual input, and more. Simulation with rendering on ManiSkill3 can run 10-1000x faster with 2-3x less GPU memory usage than other platforms, achieving up to 30,000+ FPS in benchmarked environments due to minimal python/pytorch overhead in the system, simulation on the GPU, and the use of the SAPIEN parallel rendering system. Tasks that used to take hours to train can now take minutes. We further provide the most comprehensive range of GPU parallelized environments/tasks spanning 12 distinct domains including but not limited to mobile manipulation for tasks such as drawing, humanoids, and dextrous manipulation in realistic scenes designed by artists or real-world digital twins. In addition, millions of demonstration frames are provided from motion planning, RL, and teleoperation. ManiSkill3 also provides a comprehensive set of baselines that span popular RL and learning-from-demonstrations algorithms.

cs.RO↗

arXiv2025

UniTEX: Universal High Fidelity Generative Texturing for 3D Shapes

We present UniTEX, a novel two-stage 3D texture generation framework to create high-quality, consistent textures for 3D assets. Existing approaches predominantly rely on UV-based inpainting to refine textures after reprojecting the generated multi-view images onto the 3D shapes, which introduces challenges related to topological ambiguity. To address this, we propose to bypass the limitations of UV mapping by operating directly in a unified 3D functional space. Specifically, we first propose that lifts texture generation into 3D space via Texture Functions (TFs)--a continuous, volumetric representation that maps any 3D point to a texture value based solely on surface proximity, independent of mesh topology. Then, we propose to predict these TFs directly from images and geometry inputs using a transformer-based Large Texturing Model (LTM). To further enhance texture quality and leverage powerful 2D priors, we develop an advanced LoRA-based strategy for efficiently adapting large-scale Diffusion Transformers (DiTs) for high-quality multi-view texture synthesis as our first stage. Extensive experiments demonstrate that UniTEX achieves superior visual quality and texture integrity compared to existing approaches, offering a generalizable and scalable solution for automated 3D texture generation. Code will available in: https://github.com/YixunLiang/UniTEX.

cs.CV↗

arXiv2025

Ambient-pressure superconductivity onset at 10 K and robust Tc under high pressure in TiNbTaN3 medium-entropy nitride

Superconductivity has been one of the focal points in medium and high-entropy alloys (MEAs-HEAs) since the first discovery of the HEA superconductor in 2014. Until now, most HEAs' superconducting transition temperature (Tc) has not exceeded 10 K. Here we report the first observation of superconductivity in a bulk medium-entropy nitride (MEN), TiNbTaN3, which shows a Tc of 10 K at ambient pressure. Notably, the electronic specific heat coefficient γ(H) exhibits nonlinear H-dependence behavior, which is similar to other well-studied multigap superconductors. Furthermore, TiNbTaN3 exhibits extraordinary pressure resilience, maintaining robust superconductivity under high-pressure conditions. Density functional theory (DFT) calculations indicate that pressure exerts a negligible impact on the electronic structures of TiNbTaN3, thereby corroborating the experimental observations. These findings not only advance our understanding of emergent phenomena in entropy-stabilized nitrides but also establish a new material platform for finding more high-Tc superconductors with combinations of 4d/5d transition metal elements and light elements, motivating further investigations into high-entropy functional ceramics for extreme environment applications.

cond-mat.supr-con↗

arXiv2025

Feedback-Driven Vision-Language Alignment with Minimal Human Supervision

Vision-language models (VLMs) have demonstrated remarkable potential in integrating visual and linguistic information, but their performance is often constrained by the need for extensive, high-quality image-text training data. Curation of these image-text pairs is both time-consuming and computationally expensive. To address this challenge, we introduce SVP (Sampling-based Visual Projection), a novel framework that enhances vision-language alignment without relying on manually curated text-image pairs or preference annotation. SVP leverages a small set of manually selected images, self-captioning and a pre-trained grounding model as a feedback mechanism to elicit latent information in VLMs. We evaluate our approach across six key areas: captioning, referring, visual question answering, multitasking, hallucination control, and object recall. Results demonstrate significant improvements, including a 14 % average improvement in captioning tasks, up to 12 % increase in object recall, and significantly reduced hallucinations, while maintaining question-answering capabilities. Using SVP, a small VLM achieves hallucination reductions similar to a model five times larger, while a VLM with initially poor referring capabilities more than doubles its performance, approaching parity with a model twice its size.

cs.CV↗