arXiv subjects

Wentao Jiang

Publications and source records attributed to Wentao Jiang.

At least 37 records · Page 2Linked to original sources

arXiv2025

Alignment Unlocks Complementarity: A Framework for Multiview Circuit Representation Learning

Multiview learning on Boolean circuits holds immense promise, as different graph-based representations offer complementary structural and semantic information. However, the vast structural heterogeneity between views, such as an And-Inverter Graph (AIG) versus an XOR-Majority Graph (XMG), poses a critical barrier to effective fusion, especially for self-supervised techniques like masked modeling. Naively applying such methods fails, as the cross-view context is perceived as noise. Our key insight is that functional alignment is a necessary precondition to unlock the power of multiview self-supervision. We introduce MixGate, a framework built on a principled training curriculum that first teaches the model a shared, function-aware representation space via an Equivalence Alignment Loss. Only then do we introduce a multiview masked modeling objective, which can now leverage the aligned views as a rich, complementary signal. Extensive experiments, including a crucial ablation study, demonstrate that our alignment-first strategy transforms masked modeling from an ineffective technique into a powerful performance driver.

cs.LG↗

arXiv2025

REX-RAG: Reasoning Exploration with Policy Correction in Retrieval-Augmented Generation

Reinforcement learning (RL) is emerging as a powerful paradigm for enabling large language models (LLMs) to perform complex reasoning tasks. Recent advances indicate that integrating RL with retrieval-augmented generation (RAG) allows LLMs to dynamically incorporate external knowledge, leading to more informed and robust decision making. However, we identify a critical challenge during policy-driven trajectory sampling: LLMs are frequently trapped in unproductive reasoning paths, which we refer to as "dead ends", committing to overconfident yet incorrect conclusions. This severely hampers exploration and undermines effective policy optimization. To address this challenge, we propose REX-RAG (Reasoning Exploration with Policy Correction in Retrieval-Augmented Generation), a novel framework that explores alternative reasoning paths while maintaining rigorous policy learning through principled distributional corrections. Our approach introduces two key innovations: (1) Mixed Sampling Strategy, which combines a novel probe sampling method with exploratory prompts to escape dead ends; and (2) Policy Correction Mechanism, which employs importance sampling to correct distribution shifts induced by mixed sampling, thereby mitigating gradient estimation bias. We evaluate it on seven question-answering benchmarks, and the experimental results show that REX-RAG achieves average performance gains of 5.1% on Qwen2.5-3B and 3.6% on Qwen2.5-7B over strong baselines, demonstrating competitive results across multiple datasets. The code is publicly available at https://github.com/MiliLab/REX-RAG.

cs.CL↗

arXiv2025

DeepCell: Self-Supervised Multiview Fusion for Circuit Representation Learning

We introduce DeepCell, a novel circuit representation learning framework that effectively integrates multiview information from both And-Inverter Graphs (AIGs) and Post-Mapping (PM) netlists. At its core, DeepCell employs a self-supervised Mask Circuit Modeling (MCM) strategy, inspired by masked language modeling, to fuse complementary circuit representations from different design stages into unified and rich embeddings. To our knowledge, DeepCell is the first framework explicitly designed for PM netlist representation learning, setting new benchmarks in both predictive accuracy and reconstruction quality. We demonstrate the practical efficacy of DeepCell by applying it to critical EDA tasks such as functional Engineering Change Orders (ECO) and technology mapping. Extensive experimental results show that DeepCell significantly surpasses state-of-the-art open-source EDA tools in efficiency and performance.

cs.LG↗

arXiv2025

ARFlow: Human Action-Reaction Flow Matching with Physical Guidance

Human action-reaction synthesis, a fundamental challenge in modeling causal human interactions, plays a critical role in applications ranging from virtual reality to social robotics. While diffusion-based models have demonstrated promising performance, they exhibit two key limitations for interaction synthesis: reliance on complex noise-to-reaction generators with intricate conditional mechanisms, and frequent physical violations in generated motions. To address these issues, we propose Action-Reaction Flow Matching (ARFlow), a novel framework that establishes direct action-to-reaction mappings, eliminating the need for complex conditional mechanisms. Our approach introduces a physical guidance mechanism specifically designed for Flow Matching (FM) that effectively prevents body penetration artifacts during sampling. Moreover, we discover the bias of traditional flow matching sampling algorithm and employ a reprojection method to revise the sampling direction of FM. To further enhance the reaction diversity, we incorporate randomness into the sampling process. Extensive experiments on NTU120, Chi3D and InterHuman datasets demonstrate that ARFlow not only outperforms existing methods in terms of Fréchet Inception Distance and motion diversity but also significantly reduces body collisions, as measured by our new Intersection Volume and Intersection Frequency metrics.

cs.CV↗

arXiv2025

Integrated phononic waveguide on thin-film lithium niobate on diamond

We demonstrate wavelength-scale phononic waveguides formed by transfer-printed thin-film lithium niobate (LN) on bulk diamond (LNOD), a material stack that combines the strong piezoelectricity of LN with the high acoustic velocity and color-center compatibility of diamond. We characterize a delay line based on a 100 micron long phononic waveguide at room and cryogenic temperatures. The total insertion loss through the device at 4 kelvin is -5.8 dB, corresponding to a >50% transducer efficiency, at a frequency of 2.8 gigahertz. Our work represents a step towards phonon-mediated hybrid quantum systems consisting of strain-sensitive color centers in diamond.

cond-mat.mes-hall↗

arXiv2025

ForgeEDA: A Comprehensive Multimodal Dataset for Advancing EDA

We introduce ForgeEDA, an open-source comprehensive circuit dataset across various categories. ForgeEDA includes diverse circuit representations such as Register Transfer Level (RTL) code, Post-mapping (PM) netlists, And-Inverter Graphs (AIGs), and placed netlists, enabling comprehensive analysis and development. We demonstrate ForgeEDA's utility by benchmarking state-of-the-art EDA algorithms on critical tasks such as Power, Performance, and Area (PPA) optimization, highlighting its ability to expose performance gaps and drive advancements. Additionally, ForgeEDA's scale and diversity facilitate the training of AI models for EDA tasks, demonstrating its potential to improve model performance and generalization. By addressing limitations in existing datasets, ForgeEDA aims to catalyze breakthroughs in modern IC design and support the next generation of innovations in EDA.

cs.AR↗

arXiv2025

Benchmarking Reasoning Robustness in Large Language Models

Despite the recent success of large language models (LLMs) in reasoning such as DeepSeek, we for the first time identify a key dilemma in reasoning robustness and generalization: significant performance degradation on novel or incomplete data, suggesting a reliance on memorized patterns rather than systematic reasoning. Our closer examination reveals four key unique limitations underlying this issue:(1) Positional bias--models favor earlier queries in multi-query inputs but answering the wrong one in the latter (e.g., GPT-4o's accuracy drops from 75.8 percent to 72.8 percent); (2) Instruction sensitivity--performance declines by 5.0 to 7.5 percent in the Qwen2.5 Series and by 5.0 percent in DeepSeek-V3 with auxiliary guidance; (3) Numerical fragility--value substitution sharply reduces accuracy (e.g., GPT-4o drops from 97.5 percent to 82.5 percent, GPT-o1-mini drops from 97.5 percent to 92.5 percent); and (4) Memory dependence--models resort to guesswork when missing critical data. These findings further highlight the reliance on heuristic recall over rigorous logical inference, demonstrating challenges in reasoning robustness. To comprehensively investigate these robustness challenges, this paper introduces a novel benchmark, termed as Math-RoB, that exploits hallucinations triggered by missing information to expose reasoning gaps. This is achieved by an instruction-based approach to generate diverse datasets that closely resemble training distributions, facilitating a holistic robustness assessment and advancing the development of more robust reasoning frameworks. Bad character(s) in field Abstract.

cs.AI↗

arXiv2025

Dynamic Parallel Tree Search for Efficient LLM Reasoning

Tree of Thoughts (ToT) enhances Large Language Model (LLM) reasoning by structuring problem-solving as a spanning tree. However, recent methods focus on search accuracy while overlooking computational efficiency. The challenges of accelerating the ToT lie in the frequent switching of reasoning focus, and the redundant exploration of suboptimal solutions. To alleviate this dilemma, we propose Dynamic Parallel Tree Search (DPTS), a novel parallelism framework that aims to dynamically optimize the reasoning path in inference. It includes the Parallelism Streamline in the generation phase to build up a flexible and adaptive parallelism with arbitrary paths by fine-grained cache management and alignment. Meanwhile, the Search and Transition Mechanism filters potential candidates to dynamically maintain the reasoning focus on more possible solutions and have less redundancy. Experiments on Qwen-2.5 and Llama-3 with Math500 and GSM8K datasets show that DPTS significantly improves efficiency by 2-4x on average while maintaining or even surpassing existing reasoning algorithms in accuracy, making ToT-based reasoning more scalable and computationally efficient.

cs.AI↗

arXiv2025

Correlated Dephasing in a Piezoelectrically Transduced Silicon Phononic Waveguide

Nanomechanical waveguides offer a multitude of applications in quantum and classical technologies. Here, we design, fabricate, and characterize a compact silicon single-mode phononic waveguide actuated by a thin-film lithium niobate piezoelectric element. Our device directly transduces between microwave frequency photons and phonons propagating in the silicon waveguide, providing a route for coupling to superconducting circuits. We probe the device at millikelvin temperatures through a superconducting microwave resonant matching cavity to reveal harmonics of the silicon waveguide and extract a piezoelectric coupling rate $g/2π= 1.1$ megahertz and a mechanical coupling rate $f/2π=5$ megahertz. Through time-domain measurements of the silicon mechanical modes, we observe energy relaxation timescales of $T_{1,\text{in}} \approx 500$ microseconds, pure dephasing timescales of $T_ϕ\approx {60}$ microseconds and dephasing dynamics that indicate the presence of an underlying frequency noise process with a non-uniform spectral distribution. We measure phase noise cross-correlations between silicon mechanical modes and observe detuning-dependent positively-correlated frequency fluctuations. Our measurements provide valuable insights into the dynamics and decoherence characteristics of hybrid piezoelectric-silicon acoustic devices, and suggest approaches for mitigating and circumventing noise processes for emerging quantum acoustic systems.

cond-mat.mes-hall↗

arXiv2025

Analyzing Infrastructure LiDAR Placement with Realistic LiDAR Simulation Library

Recently, Vehicle-to-Everything(V2X) cooperative perception has attracted increasing attention. Infrastructure sensors play a critical role in this research field; however, how to find the optimal placement of infrastructure sensors is rarely studied. In this paper, we investigate the problem of infrastructure sensor placement and propose a pipeline that can efficiently and effectively find optimal installation positions for infrastructure sensors in a realistic simulated environment. To better simulate and evaluate LiDAR placement, we establish a Realistic LiDAR Simulation library that can simulate the unique characteristics of different popular LiDARs and produce high-fidelity LiDAR point clouds in the CARLA simulator. Through simulating point cloud data in different LiDAR placements, we can evaluate the perception accuracy of these placements using multiple detection models. Then, we analyze the correlation between the point cloud distribution and perception accuracy by calculating the density and uniformity of regions of interest. Experiments show that when using the same number and type of LiDAR, the placement scheme optimized by our proposed method improves the average precision by 15%, compared with the conventional placement scheme in the standard lane scene. We also analyze the correlation between perception performance in the region of interest and LiDAR point cloud distribution and validate that density and uniformity can be indicators of performance. Both the RLS Library and related code will be released at https://github.com/PJLab-ADG/PCSim.

cs.RO↗

arXiv2024

Proposal for Superconducting Quantum Networks Using Multi-Octave Transduction to Lower Frequencies

We propose networking superconducting quantum circuits by transducing their excitations (typically 4-8 GHz) to 100-500 MHz photons for transmission via cryogenic coaxial cables. Counter-intuitively, this frequency downconversion reduces noise and transmission losses. We introduce a multi-octave asymmetrically threaded SQUID circuit (MOATS) capable of the required efficient, high-rate transduction. For a 100-meter cable with $Q_i = 10^5$ at 10 mK, our approach achieves single-photon fidelities of 0.962 at 200 MHz versus 0.772 at 8 GHz, and triples the lower bound on quantum channel capacity. This method enables kilometer-scale quantum links while maintaining high fidelities, combining improved performance with the practical advantages of flexible, compact coaxial cables.

quant-ph↗

arXiv2024

Quantum limits of superconducting-photonic links and their extension to mm-waves

Photonic addressing of superconducting circuits has been proposed to overcome wiring complexity and heat load challenges, but superconducting-photonic links suffer from an efficiency-noise trade-off that limits scalability. This trade-off arises because increasing power conversion efficiency requires reducing optical power, which makes the converted signal susceptible to shot noise. We analyze this trade-off and find the infidelity of qubit gates driven by photonic signals scales inversely with the number of photons used, and therefore the power efficiency of the converter. While methods like nonlinear detection or squeezed light could mitigate this effect, we consider generating higher frequency electrical signals, such as millimeter-waves (100 GHz), using laser light. At these higher frequencies, circuits have higher operating temperatures and cooling power budgets. We demonstrate an optically-driven cryogenic millimeter-wave source with a power efficiency of $10^{-4}$ that can generate ${1}~\mathrm{μW}$ of RF power at 80 GHz with 1500 thermal photons of added noise at 4 K. Using this source, we perform frequency-domain spectroscopy of superconducting NbTiN resonators at 80-90 GHz. Our results show a promising approach to alleviate the efficiency-noise constraints on optically-driven superconducting circuits while leveraging the benefits of photonic signal delivery. Further optimization of power efficiency and noise at high frequencies could make photonic control of superconducting qubits viable at temperatures exceeding 1 kelvin.

quant-ph↗

arXiv2024

A two-dimensional optomechanical crystal for quantum transduction

Integrated optomechanical systems are one of the leading platforms for manipulating, sensing, and distributing quantum information. The temperature increase due to residual optical absorption sets the ultimate limit on performance for these applications. In this work, we demonstrate a two-dimensional optomechanical crystal geometry, named \textbf{b-dagger}, that alleviates this problem through increased thermal anchoring to the surrounding material. Our mechanical mode operates at 7.4 GHz, well within the operation range of standard cryogenic microwave hardware and piezoelectric transducers. The enhanced thermalization combined with the large optomechanical coupling rates, $g_0/2π\approx 880~\mathrm{kHz}$, and high optical quality factors, $Q_\text{opt} = 2.4 \times 10^5$, enables the ground-state cooling of the acoustic mode to phononic occupancies as low as $n_\text{m} = 0.35$ from an initial temperature of 3 kelvin, as well as entering the optomechanical strong-coupling regime. Finally, we perform pulsed sideband asymmetry of our devices at a temperature below 10 millikelvin and demonstrate ground-state operation ($n_\text{m} < 0.45$) for repetition rates as high as 3 MHz. Our results extend the boundaries of optomechanical system capabilities and establish a robust foundation for the next generation of microwave-to-optical transducers with entanglement rates overcoming the decoherence rates of state-of-the-art superconducting qubits.

quant-ph↗

arXiv2024

LeMeViT: Efficient Vision Transformer with Learnable Meta Tokens for Remote Sensing Image Interpretation

Due to spatial redundancy in remote sensing images, sparse tokens containing rich information are usually involved in self-attention (SA) to reduce the overall token numbers within the calculation, avoiding the high computational cost issue in Vision Transformers. However, such methods usually obtain sparse tokens by hand-crafted or parallel-unfriendly designs, posing a challenge to reach a better balance between efficiency and performance. Different from them, this paper proposes to use learnable meta tokens to formulate sparse tokens, which effectively learn key information meanwhile improving the inference speed. Technically, the meta tokens are first initialized from image tokens via cross-attention. Then, we propose Dual Cross-Attention (DCA) to promote information exchange between image tokens and meta tokens, where they serve as query and key (value) tokens alternatively in a dual-branch structure, significantly reducing the computational complexity compared to self-attention. By employing DCA in the early stages with dense visual tokens, we obtain the hierarchical architecture LeMeViT with various sizes. Experimental results in classification and dense prediction tasks show that LeMeViT has a significant $1.7 \times$ speedup, fewer parameters, and competitive performance compared to the baseline models, and achieves a better trade-off between efficiency and performance.

cs.CV↗

arXiv2024

Data Augmentation in Human-Centric Vision

This survey presents a comprehensive analysis of data augmentation techniques in human-centric vision tasks, a first of its kind in the field. It delves into a wide range of research areas including person ReID, human parsing, human pose estimation, and pedestrian detection, addressing the significant challenges posed by overfitting and limited training data in these domains. Our work categorizes data augmentation methods into two main types: data generation and data perturbation. Data generation covers techniques like graphic engine-based generation, generative model-based generation, and data recombination, while data perturbation is divided into image-level and human-level perturbations. Each method is tailored to the unique requirements of human-centric tasks, with some applicable across multiple areas. Our contributions include an extensive literature review, providing deep insights into the influence of these augmentation techniques in human-centric vision and highlighting the nuances of each method. We also discuss open issues and future directions, such as the integration of advanced generative models like Latent Diffusion Models, for creating more realistic and diverse training data. This survey not only encapsulates the current state of data augmentation in human-centric vision but also charts a course for future research, aiming to develop more robust, accurate, and efficient human-centric vision systems.

cs.CV↗

arXiv2023

Realistic Rainy Weather Simulation for LiDARs in CARLA Simulator

Employing data augmentation methods to enhance perception performance in adverse weather has attracted considerable attention recently. Most of the LiDAR augmentation methods post-process the existing dataset by physics-based models or machine-learning methods. However, due to the limited environmental annotations and the fixed vehicle trajectories in the existing dataset, it is challenging to edit the scene and expand the diversity of traffic flow and scenario. To this end, we propose a simulator-based physical modeling approach to augment LiDAR data in rainy weather in order to improve the perception performance of LiDAR in this scenario. We complete the modeling task of the rainy weather in the CARLA simulator and establish a pipeline for LiDAR data collection. In particular, we pay special attention to the spray and splash rolled up by the wheels of surrounding vehicles in rain and complete the simulation of this special scenario through the Spray Emitter method we developed. In addition, we examine the influence of different weather conditions on the intensity of the LiDAR echo, develop a prediction network for the intensity of the LiDAR echo, and complete the simulation of 4-feat LiDAR point cloud data. In the experiment, we observe that the model augmented by the synthetic data improves the object detection task's performance in the rainy sequence of the Waymo Open Dataset. Both the code and the dataset will be made publicly available at https://github.com/PJLab-ADG/PCSim#rainypcsim.

cs.RO↗

arXiv2023

DUSA: Decoupled Unsupervised Sim2Real Adaptation for Vehicle-to-Everything Collaborative Perception

Vehicle-to-Everything (V2X) collaborative perception is crucial for autonomous driving. However, achieving high-precision V2X perception requires a significant amount of annotated real-world data, which can always be expensive and hard to acquire. Simulated data have raised much attention since they can be massively produced at an extremely low cost. Nevertheless, the significant domain gap between simulated and real-world data, including differences in sensor type, reflectance patterns, and road surroundings, often leads to poor performance of models trained on simulated data when evaluated on real-world data. In addition, there remains a domain gap between real-world collaborative agents, e.g. different types of sensors may be installed on autonomous vehicles and roadside infrastructures with different extrinsics, further increasing the difficulty of sim2real generalization. To take full advantage of simulated data, we present a new unsupervised sim2real domain adaptation method for V2X collaborative detection named Decoupled Unsupervised Sim2Real Adaptation (DUSA). Our new method decouples the V2X collaborative sim2real domain adaptation problem into two sub-problems: sim2real adaptation and inter-agent adaptation. For sim2real adaptation, we design a Location-adaptive Sim2Real Adapter (LSA) module to adaptively aggregate features from critical locations of the feature map and align the features between simulated data and real-world data via a sim/real discriminator on the aggregated global feature. For inter-agent adaptation, we further devise a Confidence-aware Inter-agent Adapter (CIA) module to align the fine-grained features from heterogeneous agents under the guidance of agent-wise confidence maps. Experiments demonstrate the effectiveness of the proposed DUSA approach on unsupervised sim2real adaptation from the simulated V2XSet dataset to the real-world DAIR-V2X-C dataset.

cs.CV↗

arXiv2023

Optimizing the Placement of Roadside LiDARs for Autonomous Driving

Multi-agent cooperative perception is an increasingly popular topic in the field of autonomous driving, where roadside LiDARs play an essential role. However, how to optimize the placement of roadside LiDARs is a crucial but often overlooked problem. This paper proposes an approach to optimize the placement of roadside LiDARs by selecting optimized positions within the scene for better perception performance. To efficiently obtain the best combination of locations, a greedy algorithm based on perceptual gain is proposed, which selects the location that can maximize the perceptual gain sequentially. We define perceptual gain as the increased perceptual capability when a new LiDAR is placed. To obtain the perception capability, we propose a perception predictor that learns to evaluate LiDAR placement using only a single point cloud frame. A dataset named Roadside-Opt is created using the CARLA simulator to facilitate research on the roadside LiDAR placement problem.

cs.CV↗