arXiv subjects

Bo Gong

Publications and source records attributed to Bo Gong.

At least 19 recordsLinked to original sources

arXiv2026

Qwen-Audio-3.0-ASR Technical Report

In recent years, automatic speech recognition (ASR) has witnessed transformative advancements driven by three complementary paradigms: data scaling, model scaling, and deep integration with large language models (LLMs). However, bridging the gap between academic benchmark performance and real-world production utility remains a persistent challenge, particularly in handling diverse regional dialects, dynamic entities and hotwords, long-range contextual information, and disfluent spontaneous speech. In this report, we present Qwen-Audio-3.0-ASR, a Mixture-of-Experts (MoE) LLM-based ASR system designed to address these production demands through a unified, instruction-following framework. The model is built upon the Qwen backbone, and is trained on tens of millions of hours of large-scale speech data. Qwen-Audio-3.0-ASR supports transcription across 30 languages and 16 Chinese dialectal varieties spanning eight major dialect regions. Beyond multilingual and dialectal recognition, the model provides production-oriented capabilities including industry-domain entity recognition, hierarchical hotword customization, native single-pass transcription polishing, and long-audio contextual modeling. We further develop a dedicated streaming variant, Qwen-Audio-3.0-ASR-Streaming, for latency-sensitive applications. Extensive evaluations on Chinese, English, multilingual, and real-world industrial test sets demonstrate state-of-the-art or highly competitive recognition performance across a broad range of evaluation conditions, with strong performance relative to leading commercial and proprietary systems including GPT-4o Transcribe and Gemini 3.1 Pro.

cs.CL↗

arXiv2026

Spurious-mode-free finite element method for scattering resonances in transmission problems

Scattering resonances arise in wave phenomena and play an important role in many applications. While extensive theoretical studies have been conducted, effective numerical computation remains limited, and most existing methods suffer from spurious modes. In this paper, we propose a spurious-mode-free method for computing scattering resonances in transmission problems. The unbounded domain is truncated using a Dirichlet-to-Neumann (DtN) map. The resonances are formulated as eigenvalues of a holomorphic Fredholm operator function, which is discretized by the finite element method. The spectrum indicator method is then used to compute the eigenvalues of the nonlinear matrix eigenvalue problems. We establish optimal order convergence and present extensive examples that demonstrate the effectiveness of the proposed method. The results are consistent with existing theoretical findings in the literature and offer new insights that may inform further theoretical developments.

math.NA↗

arXiv2026

Momentum-constrained Hybrid Heuristic Trajectory Optimization Framework with Residual-enhanced DRL for Visually Impaired Scenarios

Safe and efficient assistive planning for visually impaired scenarios remains challenging, since existing methods struggle with multi-objective optimization, generalization, and interpretability. In response, this paper proposes a Momentum-Constrained Hybrid Heuristic Trajectory Optimization Framework (MHHTOF). To balance multiple objectives of comfort and safety, the framework designs a Heuristic Trajectory Sampling Cluster (HTSC) with a Momentum-Constrained Trajectory Optimization (MTO), which suppresses abrupt velocity and acceleration changes. In addition, a novel residual-enhanced deep reinforcement learning (DRL) module refines candidate trajectories, advancing temporal modeling and policy generalization. Finally, a dual-stage cost modeling mechanism (DCMM) is introduced to regulate optimization, where costs in the Frenet space ensure consistency, and reward-driven adaptive weights in the Cartesian space integrate user preferences for interpretability and user-centric decision-making. Experimental results show that the proposed framework converges in nearly half the iterations of baselines and achieves lower and more stable costs. In complex dynamic scenarios, MHHTOF further demonstrates stable velocity and acceleration curves with reduced risk, confirming its advantages in robustness, safety, and efficiency.

cs.RO↗

arXiv2026

Edge element DtN method for electromagnetic scattering poles of perfectly conducting obstacles

Meromorphic continuation of the scattering operator leads to scattering poles (resonances) in the complex plane. Despite their significance, numerical investigation of scattering poles remains limited. In this paper, we propose and analyze a numerical method to compute electromagnetic poles of perfectly conducting obstacles. The unbounded domain for the scattering problem is truncated using the DtN mapping and the poles are shown to be the eigenvalues of a holomorphic Fredholm operator function related to Maxwell's equations. Edge elements are used for discretization. The convergence is proved using the abstract approximation theory for eigenvalue problems of holomorphic Fredholm operator functions. The proposed finite element DtN approach is free of non-physical poles. A spectral indicator method is then employed to compute the resulting nonlinear matrix eigenvalue problem. Numerical examples are presented to demonstrate the effectiveness of the method.

math.NA↗

arXiv2026

RPIQ: Residual-Projected Multi-Collaboration Closed-Loop and Single Instance Quantization for Visually Impaired Assistance

Visually impaired users face significant challenges in daily information access and real-time environmental perception, and there is an urgent need for intelligent assistive systems with accurate recognition capabilities. Although large-scale models provide effective solutions for perception and reasoning, their practical deployment on assistive devices is severely constrained by excessive memory consumption and high inference costs. Moreover, existing quantization strategies often ignore inter-block error accumulation, leading to degraded model stability. To address these challenges, this study proposes a novel quantization framework -- Residual-Projected Multi-Collaboration Closed-Loop and Single Instance Quantization(RPIQ), whose quantization process adopts a multi-collaborative closed-loop compensation scheme based on Single Instance Calibration and Gauss-Seidel Iterative Quantization. Experiments on various types of large-scale models, including language models such as OPT, Qwen, and LLaMA, as well as vision-language models such as CogVLM2, demonstrate that RPIQ can compress models to 4-bit representation while significantly reducing peak memory consumption (approximately 60%-75% reduction compared to original full-precision models). The method maintains performance highly close to full-precision models across multiple language and visual tasks, and exhibits excellent recognition and reasoning capabilities in key applications such as text understanding and visual question answering in complex scenarios. While verifying the effectiveness of RPIQ for deployment in real assistive systems, this study also advances the computational efficiency and reliability of large models, enabling them to provide visually impaired users with the required information accurately and rapidly.

cs.LG↗

arXiv2025

Momentum-constrained Hybrid Heuristic Trajectory Optimization Framework with Residual-enhanced DRL for Visually Impaired Scenarios

This paper proposes a momentum-constrained hybrid heuristic trajectory optimization framework (MHHTOF) tailored for assistive navigation in visually impaired scenarios, integrating trajectory sampling generation, optimization and evaluation with residual-enhanced deep reinforcement learning (DRL). In the first stage, heuristic trajectory sampling cluster (HTSC) is generated in the Frenet coordinate system using third-order interpolation with fifth-order polynomials and momentum-constrained trajectory optimization (MTO) constraints to ensure smoothness and feasibility. After first stage cost evaluation, the second stage leverages a residual-enhanced actor-critic network with LSTM-based temporal feature modeling to adaptively refine trajectory selection in the Cartesian coordinate system. A dual-stage cost modeling mechanism (DCMM) with weight transfer aligns semantic priorities across stages, supporting human-centered optimization. Experimental results demonstrate that the proposed LSTM-ResB-PPO achieves significantly faster convergence, attaining stable policy performance in approximately half the training iterations required by the PPO baseline, while simultaneously enhancing both reward outcomes and training stability. Compared to baseline method, the selected model reduces average cost and cost variance by 30.3% and 53.3%, and lowers ego and obstacle risks by over 77%. These findings validate the framework's effectiveness in enhancing robustness, safety, and real-time feasibility in complex assistive planning tasks.

cs.RO↗

arXiv2025

Fun-ASR Technical Report

In recent years, automatic speech recognition (ASR) has witnessed transformative advancements driven by three complementary paradigms: data scaling, model size scaling, and deep integration with large language models (LLMs). However, LLMs are prone to hallucination, which can significantly degrade user experience in real-world ASR applications. In this paper, we present Fun-ASR, a large-scale, LLM-based ASR system that synergistically combines massive data, large model capacity, LLM integration, and reinforcement learning to achieve state-of-the-art performance across diverse and complex speech recognition scenarios. Moreover, Fun-ASR is specifically optimized for practical deployment, with enhancements in streaming capability, noise robustness, code-switching, hotword customization, and satisfying other real-world application requirements. Experimental results show that while most LLM-based ASR systems achieve strong performance on open-source benchmarks, they often underperform on real industry evaluation sets. Thanks to production-oriented optimizations, Fun-ASR achieves state-of-the-art performance on real application datasets, demonstrating its effectiveness and robustness in practical settings. The code and models are accessible at https://github.com/FunAudioLLM/Fun-ASR .

cs.CL↗

arXiv2025

Development of Reduced Feeder and Load Models Using Practical Topological and Loading Data

Distribution feeder and load model reduction methods are essential for maintaining a good tradeoff between accurate representation of grid behavior and reduced computational complexity in power system studies. An effective algorithm to obtain a reduced order representation of the practical feeders using utility topological and loading data has been presented in this paper. Simulations conducted in this work show that the reduced feeder and load model of a utility feeder, obtained using the proposed method, can accurately capture contactor and motor stalling behaviors for critical events such as fault induced delayed voltage recovery.

eess.SY↗

arXiv2025

FEM-DtN-SIM Method for Computing Resonances of Schr\"{o}dinger Operators

The study of resonances of the Schr\"{o}dinger operator has a long-standing tradition in mathematical physics. Extensive theoretical investigations have explored the proximity of resonances to the real axis, their distribution, and bounds on the counting functions. However, computational results beyond one dimension remain scarce due to the nonlinearity of the problem and the unbounded nature of the domain. We propose a novel approach that integrates finite elements, Dirichlet-to-Neumann (DtN) mapping, and the spectral indicator method. The DtN mapping, imposed on the boundary of a truncated computational domain, enforces the outgoing condition. Finite elements allow for the efficient handling of complicated potential functions. The spectral indicator method effectively computes (complex) eigenvalues of the resulting nonlinear algebraic system without introducing spectral pollution. The viability of this approach is demonstrated through a range of numerical examples.

math.NA↗

arXiv2024

Analysis of a finite element DtN method for scattering resonances of sound hard obstacles

Scattering resonances have important applications in many areas of science and engineering. They are the replacement of discrete spectral data for problems on non-compact domains. In this paper, we consider the computation of scattering resonances defined on the exterior to a compact sound hard obstacle. The resonances are the eigenvalues of a holomorphic Fredholm operator function. We truncate the unbounded domain and impose the Dirichlet-to-Neumann (DtN) mapping. The problem is then discretized using the linear Lagrange element. Convergence of the resonances is proved using the abstract approximation theory for holomorphic Fredholm operator functions. The discretization leads to nonlinear algebraic eigenvalue problems, which are solved by the recently developed parallel spectral indicator methods. Numerical examples are presented for validation.

math.NA↗

arXiv2022

Regular Convergence and Finite Element Methods for Eigenvalue Problems

Regular convergence, together with various other types of convergence, has been studied since the 1970s for the discrete approximations of linear operators. In this paper, we consider the eigenvalue approximation of compact operators whose spectral problem can be written as the eigenvalue problem of some holomophic Fredholm operator function. Focusing on the finite element methods (conforming, discontinuous Galerkin, etc.), we show that the regular convergence of discrete holomorphic operator functions follows from the approximation property of the finite element spaces and the compact convergence of the discrete operators in some suitable Sobolev space. The convergence for eigenvalues is then obtained using the discrete approximation theory for the eigenvalue problems of holomorphic Fredholm operator functions. The result can be used to show the convergence of various finite element methods for eigenvalue problems such as the Dirhcilet eigenvalue problem and the biharmonic eigenvalue problem.

math.NA↗

arXiv2020

Finite Element Calculation of Photonic Band Structures for Frequency Dependent Materials

We consider the calculation of the band structure of frequency dependent photonic crystals. The associated eigenvalue problem is nonlinear and it is challenging to develop effective convergent numerical methods. In this paper, the band structure problem is formulated as the eigenvalue problem of a holomorphic Fredholm operator function of index zero. Lagrange finite elements are used to discretize the operator function. Then the convergence of the eigenvalues is proved using the abstract approximation theory for holomorphic operator functions. A spectral indicator method is developed to practically compute the eigenvalues. Numerical examples are presented to validate the theory and show the effectiveness of the proposed method.

math.NA↗

arXiv2020

SegAttnGAN: Text to Image Generation with Segmentation Attention

In this paper, we propose a novel generative network (SegAttnGAN) that utilizes additional segmentation information for the text-to-image synthesis task. As the segmentation data introduced to the model provides useful guidance on the generator training, the proposed model can generate images with better realism quality and higher quantitative measures compared with the previous state-of-art methods. We achieved Inception Score of 4.84 on the CUB dataset and 3.52 on the Oxford-102 dataset. Besides, we tested the self-attention SegAttnGAN which uses generated segmentation data instead of masks from datasets for attention and achieved similar high-quality results, suggesting that our model can be adapted for the text-to-image synthesis task.

cs.CV↗

arXiv2020

Structured Distributed Compressive Channel Estimation over Doubly Selective Channels

For an orthogonal frequency-division multiplexing (OFDM) system over a doubly selective (DS) channel, a large number of pilot subcarriers are needed to estimate the numerous channel parameters, resulting in low spectral efficiency. In this paper, by exploiting temporal correlation of practical wireless channels, we propose a highly efficient structured distributed compressive sensing (SDCS) based joint multi-symbol channel estimation scheme. Specifically, by using the complex exponential basis expansion model (CE-BEM) and exploiting the sparsity in the delay domain within multiple OFDM symbols, we turn to estimate jointly sparse CE-BEM coefficient vectors rather than numerous channel taps. Then a sparse pilot pattern within multiple OFDM symbols is designed to obtain an ICI-free structure and transform the channel estimation problem into a joint-block-sparse model. Next, a novel block-based simultaneous orthogonal matching pursuit (BSOMP) algorithm is proposed to jointly recover coefficient vectors accurately. Finally, to reduce the CE-BEM modeling error, we carry out smoothing treatments of already estimated channel taps via piecewise linear approximation.Simulation results demonstrate that the proposed channel estimation scheme can achieve higher estimation accuracy than conventional schemes, although with a smaller number of pilot subcarriers.

eess.SP↗

arXiv2020

Block Distributed Compressive Sensing Based Doubly Selective Channel Estimation and Pilot Design for Large-Scale MIMO Systems

The doubly selective (DS) channel estimation in the large-scale multiple-input multiple-output (MIMO) systems is a challenging problem due to the large number of the channel coefficients to be estimated, which requires unaffordable and prohibitive pilot overhead. In this paper, firstly we conduct the analysis about the common sparsity of the basis expansion model (BEM) coefficients among all the BEM orders and all the transmit-receive antenna pairs. Then a novel pilot pattern is proposed, which inserts the guard pilots to deal with the inter carrier interference (ICI) under the superimposed pilot pattern. Moreover, by exploiting the common sparsity of the BEM coefficients among different BEM orders and different antennas, we propose a block distributed compressive sensing (BDCS) based DS channel estimator for the large-scale MIMO systems. Its structured sparsity leads to the reduction of the pilot overhead under the premise of guaranteeing the accuracy of the estimation. Furthermore, taking consideration of the block structure, a pilot design algorithm referred to as block discrete stochastic optimization (BDSO) is proposed. It optimizes the pilot positions by reducing the coherence among different blocks of the measurement matrix. Besides, a linear smoothing method is extended to large-scale MIMO systems to improve the accuracy of the estimation. Simulation results verify the performance gains of our proposed estimator and the pilot design algorithm compared with the existing schemes.

eess.SP↗

arXiv2020

Finite Element Approximation of the Modified Maxwell's Stekloff Eigenvalues

The modified Maxwell's Stekloff eigenvalue problem arises recently from the inverse electromagnetic scattering theory for inhomogeneous media. This paper contains a rigorous analysis of both the eigenvalue problem and the associated source problem on Lipschitz polyhedra. A new finite element method is proposed to compute Stekloff eigenvalues. By applying the Babuska-Osborn theory, we prove an error estimate without additional regularity assumptions. Numerical results are presented for validation.

math.NA↗

arXiv2020

Position-Based Interference Elimination for High Mobility OFDM Channel Estimation in Multi-cell Systems

Orthogonal frequency-division multiplexing (OFD-M) and multi-cell architecture are widely adopted in current high speed train (HST) systems for providing high data rate wireless communications. In this paper, a typical multi-antenna OFDM HST communication system with multi-cell architecture is considered, where the inter-carrier interference (ICI) caused by high mobility and multi-cell interference (MCI) are both taken into consideration. By exploiting the train position information, a new position-based interference elimination method is proposed to eliminate both the MCI and ICI for a general basis expansion model (BEM). We show that the MCI and ICI can be completely eliminated by the proposed method to get the ICI-free pilots at each receive antenna. In addition, for the considered multi-cell HST system, we develop a low-complexity compressed channel estimation method and consider the optimal pilot pattern design. Both the proposed interference elimination method and the optimal pilot pattern are robust to the train speed and position,as well as the multi-cell multi-antenna system. Simulation results demonstrate the benefits and robustness of the proposed method in the multi-cell HST system.

eess.SP↗

arXiv2020

A new finite element approach for the Dirichlet eigenvalue problem

In this paper, we propose a new finite element approach, which is different than the classic Babuska-Osborn theory, to approximate Dirichlet eigenvalues. The Dirichlet eigenvalue problem is formulated as the eigenvalue problem of a holomorphic Fredholm operator function of index zero. Using conforming finite elements, the convergence is proved using the abstract approximation theory for holomorphic operator functions. The spectral indicator method is employed to compute the eigenvalues. A numerical example is presented to validate the theory.

math.NA↗