arXiv subjects

Gang Wang

Publications and source records attributed to Gang Wang.

At least 109 records · Page 6Linked to original sources

arXiv2025

A Chaotic Dynamics Framework Inspired by Dorsal Stream for Event Signal Processing

Event cameras are bio-inspired vision sensor that encode visual information with high dynamic range, high temporal resolution, and low latency.Current state-of-the-art event stream processing methods rely on end-to-end deep learning techniques. However, these models are heavily dependent on data structures, limiting their stability and generalization capabilities across tasks, thereby hindering their deployment in real-world scenarios. To address this issue, we propose a chaotic dynamics event signal processing framework inspired by the dorsal visual pathway of the brain. Specifically, we utilize Continuous-coupled Neural Network (CCNN) to encode the event stream. CCNN encodes polarity-invariant event sequences as periodic signals and polarity=changing event sequences as chaotic signals. We then use continuous wavelet transforms to analyze the dynamical states of CCNN neurons and establish the high-order mappings of the event stream. The effectiveness of our method is validated through integration with conventional classification networks, achieving state-of-the-art classification accuracy on the N-Caltech101 and N-CARS datasets, with results of 84.3% and 99.9%, respectively. Our method improves the accuracy of event camera-based object classification while significantly enhancing the generalization and stability of event representation. Our code is available in https://github.com/chenyu0193/ACDF.

q-bio.NC↗

arXiv2025

DyMoDreamer: World Modeling with Dynamic Modulation

A critical bottleneck in deep reinforcement learning (DRL) is sample inefficiency, as training high-performance agents often demands extensive environmental interactions. Model-based reinforcement learning (MBRL) mitigates this by building world models that simulate environmental dynamics and generate synthetic experience, improving sample efficiency. However, conventional world models process observations holistically, failing to decouple dynamic objects and temporal features from static backgrounds. This approach is computationally inefficient, especially for visual tasks where dynamic objects significantly influence rewards and decision-making performance. To address this, we introduce DyMoDreamer, a novel MBRL algorithm that incorporates a dynamic modulation mechanism to improve the extraction of dynamic features and enrich the temporal information. DyMoDreamer employs differential observations derived from a novel inter-frame differencing mask, explicitly encoding object-level motion cues and temporal dynamics. Dynamic modulation is modeled as stochastic categorical distributions and integrated into a recurrent state-space model (RSSM), enhancing the model's focus on reward-relevant dynamics. Experiments demonstrate that DyMoDreamer sets a new state-of-the-art on the Atari $100$k benchmark with a $156.6$\% mean human-normalized score, establishes a new record of $832$ on the DeepMind Visual Control Suite, and gains a $9.5$\% performance improvement after $1$M steps on the Crafter benchmark. Our code is released at https://github.com/Ultraman-Tiga1/DyMoDreamer.

cs.LG↗

arXiv2025

Detectability of massive binary black holes with sub-mHz gravitational wave missions

Beyond LISA, proposed space-based gravitational wave (GW) missions aim to explore the sub-millihertz to microhertz frequency band, with one key objective being the detection of massive binary black hole (MBBH) mergers across cosmic distances. In this work, we investigate the detection and localization capabilities of future sub-mHz GW observatories for MBBH coalescences. Including the full galactic foreground noise, we find that signal-to-noise ratios (SNRs) can reach several thousand across a wide range of redshifts. We evaluate three representative orbital configurations--non-precessing and precessing with different inclination angles--and analyze their localization performance for various MBBH populations. In the non-precessing case, a two-hemisphere degeneracy arises when only the dominant (2,2) mode is considered, which is effectively resolved by including higher-order modes. These modes contribute to a more uniform performance across all configurations, thereby mitigating the prior advantage of precessing mission orbits. Sub-mHz missions operating in the [10 $μ$Hz, 10 mHz] band partially overlap with LISA's range but provide enhanced sensitivity to lower-frequency GWs due to their longer interferometric baselines. This results in significantly improved localization of high-mass MBBHs, enhancing the prospects for multi-messenger astronomy and precision cosmology. Moreover, the high SNRs attainable with sub-mHz detectors could enable stringent tests of general relativity and alternative theories of gravity.

gr-qc↗

arXiv2025

New constructions of cyclic constant-dimension subspace codes based on Sidon spaces and subspace polynomials

In this paper, two new constructions of Sidon spaces are given by tactfully adding new parameters and flexibly varying the number of parameters. Under the parameters $ n= (2r+1)k, r \ge2 $ and $p_0=\max \{i\in \mathbb{N}^+: \lfloor \frac{r}{i}\rfloor>\lfloor \frac{r}{i+1} \rfloor \}$, the first construction produces a cyclic CDC in $\mathcal{G}_q(n, k)$ with minimum distance $2k-2$ and size $\frac{\left((r+\sum\limits_{i=2}^{p_0}(\lfloor \frac{r}{i}\rfloor-\lfloor \frac{r}{i+1} \rfloor))(q^k-1)(q-1)+r\right)(q^k-1)^{r-1}(q^n-1)}{q-1}$. Given parameters $n=2rk,r\ge 2$ and if $r=2$, $p_0=1$, otherwise, $p_0=\max\{ i\in \mathbb{N}^+: \lceil\frac{r}{i}\rceil-1>\lfloor \frac{r}{i+1} \rfloor \}$, a cyclic CDC in $\mathcal{G}_q(n, k)$ with minimum distance $2k-2$ and size $\frac{\left((r-1+\sum\limits_{i=2}^{p_0}(\lceil \frac{r}{i}\rceil-\lfloor \frac{r}{i+1} \rfloor-1))(q^k-1)(q-1)+r-1\right)(q^k-1)^{r-2}\lfloor \frac{q^k-2}{2}\rfloor(q^n-1)}{q-1}$ is produced by the second construction. The sizes of our cyclic CDCs are larger than the best known results. In particular, in the case of $n=4k$, when $k$ goes to infinity, the ratio between the size of our cyclic CDC and the Sphere-packing bound (Johnson bound) is approximately equal to $\frac{1}{2}$. Moreover, for a prime power $q$ and positive integers $k,s$ with $1\le s< k-1$, a cyclic CDC in $\mathcal{G}_q(N, k)$ of size $e\frac{q^N-1}{q-1}$ and minimum distance $\ge 2k-2s$ is provided by subspace polynomials, where $N,e$ are positive integers. Our construction generalizes previous results and, under certain parameters, provides cyclic CDCs with larger sizes or more admissible values of $ N $ than constructions based on trinomials.

cs.IT↗

arXiv2025

HoloDx: Knowledge- and Data-Driven Multimodal Diagnosis of Alzheimer's Disease

Accurate diagnosis of Alzheimer's disease (AD) requires effectively integrating multimodal data and clinical expertise. However, existing methods often struggle to fully utilize multimodal information and lack structured mechanisms to incorporate dynamic domain knowledge. To address these limitations, we propose HoloDx, a knowledge- and data-driven framework that enhances AD diagnosis by aligning domain knowledge with multimodal clinical data. HoloDx incorporates a knowledge injection module with a knowledge-aware gated cross-attention, allowing the model to dynamically integrate domain-specific insights from both large language models (LLMs) and clinical expertise. Also, a memory injection module with a designed prototypical memory attention enables the model to retain and retrieve subject-specific information, ensuring consistency in decision-making. By jointly leveraging these mechanisms, HoloDx enhances interpretability, improves robustness, and effectively aligns prior knowledge with current subject data. Evaluations on five AD datasets demonstrate that HoloDx outperforms state-of-the-art methods, achieving superior diagnostic accuracy and strong generalization across diverse cohorts. The source code will be released upon publication acceptance.

cs.CV↗

arXiv2025

OneRec Technical Report

Recommender systems have been widely used in various large-scale user-oriented platforms for many years. However, compared to the rapid developments in the AI community, recommendation systems have not achieved a breakthrough in recent years. For instance, they still rely on a multi-stage cascaded architecture rather than an end-to-end approach, leading to computational fragmentation and optimization inconsistencies, and hindering the effective application of key breakthrough technologies from the AI community in recommendation scenarios. To address these issues, we propose OneRec, which reshapes the recommendation system through an end-to-end generative approach and achieves promising results. Firstly, we have enhanced the computational FLOPs of the current recommendation model by 10 $\times$ and have identified the scaling laws for recommendations within certain boundaries. Secondly, reinforcement learning techniques, previously difficult to apply for optimizing recommendations, show significant potential in this framework. Lastly, through infrastructure optimizations, we have achieved 23.7% and 28.8% Model FLOPs Utilization (MFU) on flagship GPUs during training and inference, respectively, aligning closely with the LLM community. This architecture significantly reduces communication and storage overhead, resulting in operating expense that is only 10.6% of traditional recommendation pipelines. Deployed in Kuaishou/Kuaishou Lite APP, it handles 25% of total queries per second, enhancing overall App Stay Time by 0.54% and 1.24%, respectively. Additionally, we have observed significant increases in metrics such as 7-day Lifetime, which is a crucial indicator of recommendation experience. We also provide practical lessons and insights derived from developing, optimizing, and maintaining a production-scale recommendation system with significant real-world impact.

cs.IR↗

arXiv2025

CORB-Planner: Corridor as Observations for RL Planning in High-Speed Flight

Reinforcement learning (RL) has shown promise in a large number of robotic control tasks. Nevertheless, its deployment on unmanned aerial vehicles (UAVs) remains challenging, mainly because of reliance on accurate dynamic models and platform-specific sensing, which hinders cross-platform transfer. This paper presents the CORB-Planner (Corridor-as-Observations for RL B-spline planner), a real-time, RL-based trajectory planning framework for high-speed autonomous UAV flight across heterogeneous platforms. The key idea is to combine B-spline trajectory generation with the RL policy producing successive control points with a compact safe flight corridor (SFC) representation obtained via heuristic search. The SFC abstracts obstacle information in a low-dimensional form, mitigating overfitting to platform-specific details and reducing sensitivity to model inaccuracies. To narrow the sim-to-real gap, we adopt an easy-to-hard progressive training pipeline in simulation. A value-based soft decomposed-critic Q (SDCQ) algorithm is used to learn effective policies within approximately ten minutes of training. Benchmarks in simulation and real-world tests demonstrate real-time planning on lightweight onboard hardware and support maximum flight speeds up to 8.2m/s in dense, cluttered environments without external positioning. Compatibility with various UAV configurations (quadrotors, hexarotors) and modest onboard compute underlines the generality and robustness of CORB-Planner for practical deployment.

cs.RO↗

arXiv2025

DiffAero: A GPU-Accelerated Differentiable Simulation Framework for Efficient Quadrotor Policy Learning

This letter introduces DiffAero, a lightweight, GPU-accelerated, and fully differentiable simulation framework designed for efficient quadrotor control policy learning. DiffAero supports both environment-level and agent-level parallelism and integrates multiple dynamics models, customizable sensor stacks (IMU, depth camera, and LiDAR), and diverse flight tasks within a unified, GPU-native training interface. By fully parallelizing both physics and rendering on the GPU, DiffAero eliminates CPU-GPU data transfer bottlenecks and delivers orders-of-magnitude improvements in simulation throughput. In contrast to existing simulators, DiffAero not only provides high-performance simulation but also serves as a research platform for exploring differentiable and hybrid learning algorithms. Extensive benchmarks and real-world flight experiments demonstrate that DiffAero and hybrid learning algorithms combined can learn robust flight policies in hours on consumer-grade hardware. The code is available at https://github.com/flyingbitac/diffaero.

cs.RO↗

arXiv2025

Robust Offline Imitation Learning Through State-level Trajectory Stitching

Imitation learning (IL) has proven effective for enabling robots to acquire visuomotor skills through expert demonstrations. However, traditional IL methods are limited by their reliance on high-quality, often scarce, expert data, and suffer from covariate shift. To address these challenges, recent advances in offline IL have incorporated suboptimal, unlabeled datasets into the training. In this paper, we propose a novel approach to enhance policy learning from mixed-quality offline datasets by leveraging task-relevant trajectory fragments and rich environmental dynamics. Specifically, we introduce a state-based search framework that stitches state-action pairs from imperfect demonstrations, generating more diverse and informative training trajectories. Experimental results on standard IL benchmarks and real-world robotic tasks showcase that our proposed method significantly improves both generalization and performance.

cs.RO↗

arXiv2025

SPGrasp: Spatiotemporal Prompt-driven Grasp Synthesis in Dynamic Scenes

Real-time interactive grasp synthesis for dynamic objects remains challenging as existing methods fail to achieve low-latency inference while maintaining promptability. To bridge this gap, we propose SPGrasp (spatiotemporal prompt-driven dynamic grasp synthesis), a novel framework extending segment anything model v2 (SAMv2) for video stream grasp estimation. Our core innovation integrates user prompts with spatiotemporal context, enabling real-time interaction with end-to-end latency as low as 59 ms while ensuring temporal consistency for dynamic objects. In benchmark evaluations, SPGrasp achieves instance-level grasp accuracies of 90.6% on OCID and 93.8% on Jacquard. On the challenging GraspNet-1Billion dataset under continuous tracking, SPGrasp achieves 92.0% accuracy with 73.1 ms per-frame latency, representing a 58.5% reduction compared to the prior state-of-the-art promptable method RoG-SAM while maintaining competitive accuracy. Real-world experiments involving 13 moving objects demonstrate a 94.8% success rate in interactive grasping scenarios. These results confirm SPGrasp effectively resolves the latency-interactivity trade-off in dynamic grasp synthesis.

cs.RO↗

arXiv2025

Deep Neural Network-Driven Adaptive Filtering

This paper proposes a deep neural network (DNN)-driven framework to address the longstanding generalization challenge in adaptive filtering (AF). In contrast to traditional AF frameworks that emphasize explicit cost function design, the proposed framework shifts the paradigm toward direct gradient acquisition. The DNN, functioning as a universal nonlinear operator, is structurally embedded into the core architecture of the AF system, establishing a direct mapping between filtering residuals and learning gradients. The maximum likelihood is adopted as the implicit cost function, rendering the derived algorithm inherently data-driven and thus endowed with exemplary generalization capability, which is validated by extensive numerical experiments across a spectrum of non-Gaussian scenarios. Corresponding mean value and mean square stability analyses are also conducted in detail.

stat.ML↗

arXiv2025

Multilevel inserting constructions for constant dimension subspace codes

Subspace codes, especially constant dimension subspace codes (CDCs), represent an intriguing domain that can be used to conduct basic coding theory investigations. They have received widespread attention due to their applications in random network coding. This paper presents inverse bilateral multilevel construction by introducing inverse bilateral identifying vectors and inverse bilateral Ferrers diagram rank-metric codes. By inserting the inverse bilateral multilevel construction into the double multilevel construction and bilateral multilevel construction, an effective construction for CDCs is provided. Furthermore, via providing a new set of bilateral identifying vectors, we give another efficient construction for CDCs. In this article, several CDCs are exhibited, equipped with the rank-metric, with larger sizes than the known ones in the existing literature. From a practical standpoint, our results could help in the pragmatic framework of constant-dimension-lifted rank-metric codes for applications in network coding. The ratio of the new lower bound to the known upper bound for some CDCs is calculated, which is greater than 0.94548 for any prime power $q \geq 3.$

cs.IT↗

arXiv2025

Interstitial oxygen order and its competition with superconductivity in La$_2$PrNi$_2$O$_{7+δ}$

High-temperature superconductivity in bilayer nickelate La$_3$Ni$_2$O$_7$ under pressure has attracted significant interest in condensed matter physics. While early samples exhibited limited superconducting volume fractions, Pr substitution for La enabled bulk superconductivity in polycrystals under pressure and enhanced transition temperatures in thin films at ambient pressure. Beyond rare-earth doping, moderate oxygen or ozone annealing improves superconductivity by mitigating oxygen vacancies, whereas high-pressure oxygen annealing leads to a trivial, non-superconducting metallic state across all pressure regimes. These findings highlight the need to elucidate both the individual and combined effects of Pr doping and oxygen stoichiometry in modulating superconductivity in bilayer nickelates. Here, using multislice electron ptychography and electron energy-loss spectroscopy, we investigate the structural and electronic properties of as-grown La$_2$PrNi$_2$O$_7$ and high-pressure-oxygen-annealed La$_2$PrNi$_2$O$_{7+δ}$ polycrystals. We find that Pr dopants preferentially occupy outer La sites, effectively eliminating inner-apical oxygen vacancies and ensuring near-stoichiometry in as-grown La$_2$PrNi$_2$O$_7$ that is bulk-superconducting under pressure. In contrast, high-pressure oxygen annealing induces a striped interstitial oxygen order, introducing quasi-1D lattice potentials and excess hole carriers into p-d hybridized orbitals, ultimately suppressing superconductivity. This behavior starkly contrasts with cuprate superconductors, where similar interstitial oxygen ordering enhances superconductivity instead. Our findings reveal a competition between striped interstitial oxygen order and superconductivity in bilayer nickelates, offering key insights into their distinct pairing mechanisms and providing a roadmap for designing more robust superconducting phases.

cond-mat.supr-con↗

arXiv2025

A Group Consensus-Driven Auction Algorithm for Cooperative Task Allocation Among Heterogeneous Multi-Agents

In scenarios like automated warehouses, assigning tasks to robots presents a heterogeneous multi-task and multi-agent task allocation problem. However, existing task allocation study ignores the integration of multi-task and multi-attribute agent task allocation with heterogeneous task allocation. In addition, current algorithms are limited by scenario constraints and can incur significant errors in specific contexts. Therefore, this study proposes a distributed heterogeneous multi-task and multi-agent task allocation algorithm with a time window, called group consensus-based heterogeneous auction (GCBHA). Firstly, this method decomposes tasks that exceed the capability of a single Agent into subtasks that can be completed by multiple independent agents. And then groups similar or adjacent tasks through a heuristic clustering method to reduce the time required to reach a consensus. Subsequently, the task groups are allocated to agents that meet the conditions through an auction process. Furthermore, the method evaluates the task path cost distance based on the scenario, which can calculate the task cost more accurately. The experimental results demonstrate that GCBHA performs well in terms of task allocation time and solution quality, with a significant reduction in the error rate between predicted task costs and actual costs.

cs.MA↗

arXiv2025

New constant-dimension subspace codes from parallel cosets of optimal Ferrers diagram rank-metric codes and multilevel inserting constructions

Constant-dimension subspace codes (CDCs), a special class of subspace codes, have attracted significant attention due to their applications in network coding. A fundamental research problem of CDCs is to determine the maximum number of codewords under the given parameters. The paper first proposes the construction of parallel cosets of optimal Ferrers diagram rank-metric codes (FDRMCs) by employing the list of CDCs and inverse list of CDCs. Then a new class of CDCs is obtained by combining the parallel cosets of optimal FDRMCs with parallel linkage construction. Next, we present a novel set of identifying vectors and provide a new construction of CDCs via the multilevel constuction. Finally, the coset construction is inserted into the multilevel construction and three classes of large CDCs are provided, one of which is constructed by using new optimal FDRMCs. Our results establish at least 65 new lower bounds for CDCs with larger sizes than the previously best known codes.

cs.IT↗

arXiv2025

Next Tokens Denoising for Speech Synthesis

While diffusion and autoregressive (AR) models have significantly advanced generative modeling, they each present distinct limitations. AR models, which rely on causal attention, cannot exploit future context and suffer from slow generation speeds. Conversely, diffusion models struggle with key-value (KV) caching. To overcome these challenges, we introduce Dragon-FM, a novel text-to-speech (TTS) design that unifies AR and flow-matching. This model processes 48 kHz audio codec tokens in chunks at a compact rate of 12.5 tokens per second. This design enables AR modeling across chunks, ensuring global coherence, while parallel flow-matching within chunks facilitates fast iterative denoising. Thus, the model leverages KV-cache across chunks and utilizes bidirectional context within each chunk. Furthermore, it bridges continuous and discrete feature modeling, demonstrating that continuous AR flow-matching can predict discrete tokens with finite scalar quantizers. This efficient codec and fast chunk-autoregressive architecture also make the model highly effective for generating long-form content, such as podcasts. Experiments on podcast datasets demonstrate its capability to efficiently generate high-quality zero-shot podcasts.

cs.SD↗

arXiv2025

Data-Driven Extended Corresponding State Approach for Residual Property Prediction of Hydrofluoroolefins

Hydrofluoroolefins are considered the most promising next-generation refrigerants due to their extremely low global warming potential values, which can effectively mitigate the global warming effect. However, the lack of reliable thermodynamic data hinders the discovery and application of newer and superior hydrofluoroolefin refrigerants. In this work, integrating the strengths of theoretical method and data-driven method, we proposed a neural network extended corresponding state model to predict the residual thermodynamic properties of hydrofluoroolefin refrigerants. The innovation is that the fluids are characterized through their microscopic molecular structures by the inclusion of graph neural network module and the specialized design of model architecture to enhance its generalization ability. The proposed model is trained using the highly accurate data of available known fluids, and evaluated via the leave-one-out cross-validation method. Compared to conventional extended corresponding state models or cubic equation of state, the proposed model shows significantly improved accuracy for density and energy properties in liquid and supercritical regions, with average absolute deviation of 1.49% (liquid) and 2.42% (supercritical) for density, 3.37% and 2.50% for residual entropy, 1.85% and 1.34% for residual enthalpy. These results demonstrate the effectiveness of embedding physics knowledge into the machine learning model. The proposed neural network extended corresponding state model is expected to significantly accelerate the discovery of novel hydrofluoroolefin refrigerants.

cs.LG↗

arXiv2025

MOCHA: Are Code Language Models Robust Against Multi-Turn Malicious Coding Prompts?

Recent advancements in Large Language Models (LLMs) have significantly enhanced their code generation capabilities. However, their robustness against adversarial misuse, particularly through multi-turn malicious coding prompts, remains underexplored. In this work, we introduce code decomposition attacks, where a malicious coding task is broken down into a series of seemingly benign subtasks across multiple conversational turns to evade safety filters. To facilitate systematic evaluation, we introduce \benchmarkname{}, a large-scale benchmark designed to evaluate the robustness of code LLMs against both single-turn and multi-turn malicious prompts. Empirical results across open- and closed-source models reveal persistent vulnerabilities, especially under multi-turn scenarios. Fine-tuning on MOCHA improves rejection rates while preserving coding ability, and importantly, enhances robustness on external adversarial datasets with up to 32.4% increase in rejection rates without any additional supervision.

cs.CL↗