arXiv subjects

Peng Wang

Publications and source records attributed to Peng Wang.

At least 325 records · Page 18Linked to original sources

arXiv2025

Remote entanglement generation via enhanced quantum state transfer

Achieving robust and scalable remote quantum entanglement is a fundamental challenge for the development of distributed quantum networks and modular quantum computing systems. Along this, perfect state transfer (PST) and fractional state transfer (FST) have emerged as promising schemes for quantum state transfer and remote entanglement generation using only nearest-neighbor couplings. However, the current implementations suffer from quantum loss and limited parameter tunability. In this work, we propose a new quantum state transfer scheme based on a zig-zag configuration, which introduces a controlling parameter for PST and FST. We show that this new parameter can suppress the population in the intermediate qubits, thereby reducing losses. We experimentally demonstrate the dynamics of different configurations on a superconducting quantum processor, achieving an $18\%$ reduction in error for remote Bell state generation in a 1D ($1\times5$) qubit chain, and exhibit robustness against certain types of noise. Then we extend our approach to a 2D network, successfully generating a W state among the four corner qubits. These results highlight the potential of our enhanced quantum state transfer scheme for scalable and noise-resilient quantum communication and computing.

quant-ph↗

arXiv2025

SeedEdit 3.0: Fast and High-Quality Generative Image Editing

We introduce SeedEdit 3.0, in companion with our T2I model Seedream 3.0, which significantly improves over our previous SeedEdit versions in both aspects of edit instruction following and image content (e.g., ID/IP) preservation on real image inputs. Additional to model upgrading with T2I, in this report, we present several key improvements. First, we develop an enhanced data curation pipeline with a meta-info paradigm and meta-info embedding strategy that help mix images from multiple data sources. This allows us to scale editing data effectively, and meta information is helpfult to connect VLM with diffusion model more closely. Second, we introduce a joint learning pipeline for computing a diffusion loss and reward losses. Finally, we evaluate SeedEdit 3.0 on our testing benchmarks, for real/synthetic image editing, where it achieves a best trade-off between multiple aspects, yielding a high usability rate of 56.1%, compared to SeedEdit 1.6 (38.4%), GPT4o (37.1%) and Gemini 2.0 (30.3%).

cs.CV↗

arXiv2025

MiMo: Unlocking the Reasoning Potential of Language Model -- From Pretraining to Posttraining

We present MiMo-7B, a large language model born for reasoning tasks, with optimization across both pre-training and post-training stages. During pre-training, we enhance the data preprocessing pipeline and employ a three-stage data mixing strategy to strengthen the base model's reasoning potential. MiMo-7B-Base is pre-trained on 25 trillion tokens, with additional Multi-Token Prediction objective for enhanced performance and accelerated inference speed. During post-training, we curate a dataset of 130K verifiable mathematics and programming problems for reinforcement learning, integrating a test-difficulty-driven code-reward scheme to alleviate sparse-reward issues and employing strategic data resampling to stabilize training. Extensive evaluations show that MiMo-7B-Base possesses exceptional reasoning potential, outperforming even much larger 32B models. The final RL-tuned model, MiMo-7B-RL, achieves superior performance on mathematics, code and general reasoning tasks, surpassing the performance of OpenAI o1-mini. The model checkpoints are available at https://github.com/xiaomimimo/MiMo.

cs.CL↗

arXiv2025

MiMo-VL Technical Report

We open-source MiMo-VL-7B-SFT and MiMo-VL-7B-RL, two powerful vision-language models delivering state-of-the-art performance in both general visual understanding and multimodal reasoning. MiMo-VL-7B-RL outperforms Qwen2.5-VL-7B on 35 out of 40 evaluated tasks, and scores 59.4 on OlympiadBench, surpassing models with up to 78B parameters. For GUI grounding applications, it sets a new standard with 56.1 on OSWorld-G, even outperforming specialized models such as UI-TARS. Our training combines four-stage pre-training (2.4 trillion tokens) with Mixed On-policy Reinforcement Learning (MORL) integrating diverse reward signals. We identify the importance of incorporating high-quality reasoning data with long Chain-of-Thought into pre-training stages, and the benefits of mixed RL despite challenges in simultaneous multi-domain optimization. We also contribute a comprehensive evaluation suite covering 50+ tasks to promote reproducibility and advance the field. The model checkpoints and full evaluation suite are available at https://github.com/XiaomiMiMo/MiMo-VL.

cs.CL↗

arXiv2025

Attention-Only Transformers via Unrolled Subspace Denoising

Despite the popularity of transformers in practice, their architectures are empirically designed and neither mathematically justified nor interpretable. Moreover, as indicated by many empirical studies, some components of transformer architectures may be redundant. To derive a fully interpretable transformer architecture with only necessary components, we contend that the goal of representation learning is to compress a set of noisy initial token representations towards a mixture of low-dimensional subspaces. To compress these noisy token representations, an associated denoising operation naturally takes the form of a multi-head (subspace) self-attention. By unrolling such iterative denoising operations into a deep network, we arrive at a highly compact architecture that consists of \textit{only} self-attention operators with skip connections at each layer. Moreover, we show that each layer performs highly efficient denoising: it improves the signal-to-noise ratio of token representations \textit{at a linear rate} with respect to the number of layers. Despite its simplicity, extensive experiments on vision and language tasks demonstrate that such a transformer achieves performance close to that of standard transformer architectures such as GPT-2 and CRATE.

cs.LG↗

arXiv2025

Classification of aut-fixed subgroups in free-abelian times surface groups

In this paper, we are concerned with the direct product $G=π_1(Σ)\times \Z^k$ for $Σ$ a compact orientable surface with negative Euler characteristic, and give a complete classification of its fixed subgroups of automorphisms. As a corollary, we show that $G$ contains, up to isomorphism, infinitely many fixed subgroups of automorphisms if and only if $k\geq 2$, which is a contrast to that of hyperbolic groups. As an application on Nielsen fixed point theory, we provide a family of aspherical manifolds without Jiang's Bound Index Property. Moreover, we also give some results on the fixed subgroups of the direct product $H\times \Z^k$ for $H$ a non-elementary torsion-free hyperbolic group.

math.GR↗

arXiv2025

AVadCLIP: Audio-Visual Collaboration for Robust Video Anomaly Detection

With the increasing adoption of video anomaly detection in intelligent surveillance domains, conventional visual-based detection approaches often struggle with information insufficiency and high false-positive rates in complex environments. To address these limitations, we present a novel weakly supervised framework that leverages audio-visual collaboration for robust video anomaly detection. Capitalizing on the exceptional cross-modal representation learning capabilities of Contrastive Language-Image Pretraining (CLIP) across visual, audio, and textual domains, our framework introduces two major innovations: an efficient audio-visual fusion that enables adaptive cross-modal integration through lightweight parametric adaptation while maintaining the frozen CLIP backbone, and a novel audio-visual prompt that dynamically enhances text embeddings with key multimodal information based on the semantic correlation between audio-visual features and textual labels, significantly improving CLIP's generalization for the video anomaly detection task. Moreover, to enhance robustness against modality deficiency during inference, we further develop an uncertainty-driven feature distillation module that synthesizes audio-visual representations from visual-only inputs. This module employs uncertainty modeling based on the diversity of audio-visual features to dynamically emphasize challenging features during the distillation process. Our framework demonstrates superior performance across multiple benchmarks, with audio integration significantly boosting anomaly detection accuracy in various scenarios. Notably, with unimodal data enhanced by uncertainty-driven distillation, our approach consistently outperforms current unimodal VAD methods.

cs.CV↗

arXiv2025

Demystifying Catastrophic Forgetting in Two-Stage Incremental Object Detector

Catastrophic forgetting is a critical chanllenge for incremental object detection (IOD). Most existing methods treat the detector monolithically, relying on instance replay or knowledge distillation without analyzing component-specific forgetting. Through dissection of Faster R-CNN, we reveal a key insight: Catastrophic forgetting is predominantly localized to the RoI Head classifier, while regressors retain robustness across incremental stages. This finding challenges conventional assumptions, motivating us to develop a framework termed NSGP-RePRE. Regional Prototype Replay (RePRE) mitigates classifier forgetting via replay of two types of prototypes: coarse prototypes represent class-wise semantic centers of RoI features, while fine-grained prototypes model intra-class variations. Null Space Gradient Projection (NSGP) is further introduced to eliminate prototype-feature misalignment by updating the feature extractor in directions orthogonal to subspace of old inputs via gradient projection, aligning RePRE with incremental learning dynamics. Our simple yet effective design allows NSGP-RePRE to achieve state-of-the-art performance on the Pascal VOC and MS COCO datasets under various settings. Our work not only advances IOD methodology but also provide pivotal insights for catastrophic forgetting mitigation in IOD. Code is available at \href{https://github.com/fanrena/NSGP-RePRE}{https://github.com/fanrena/NSGP-RePRE} .

cs.CV↗

arXiv2025

SLearnLLM: A Self-Learning Framework for Efficient Domain-Specific Adaptation of Large Language Models

When using supervised fine-tuning (SFT) to adapt large language models (LLMs) to specific domains, a significant challenge arises: should we use the entire SFT dataset for fine-tuning? Common practice often involves fine-tuning directly on the entire dataset due to limited information on the LLM's past training data. However, if the SFT dataset largely overlaps with the model's existing knowledge, the performance gains are minimal, leading to wasted computational resources. Identifying the unknown knowledge within the SFT dataset and using it to fine-tune the model could substantially improve the training efficiency. To address this challenge, we propose a self-learning framework for LLMs inspired by human learning pattern. This framework takes a fine-tuning (SFT) dataset in a specific domain as input. First, the LLMs answer the questions in the SFT dataset. The LLMs then objectively grade the responses and filter out the incorrectly answered QA pairs. Finally, we fine-tune the LLMs based on this filtered QA set. Experimental results in the fields of agriculture and medicine demonstrate that our method substantially reduces training time while achieving comparable improvements to those attained with full dataset fine-tuning. By concentrating on the unknown knowledge within the SFT dataset, our approach enhances the efficiency of fine-tuning LLMs.

cs.CL↗

arXiv2025

Privacy-Preserving Socialized Recommendation based on Multi-View Clustering in a Cloud Environment

Recommendation as a service has improved the quality of our lives and plays a significant role in variant aspects. However, the preference of users may reveal some sensitive information, so that the protection of privacy is required. In this paper, we propose a privacy-preserving, socialized, recommendation protocol that introduces information collected from online social networks to enhance the quality of the recommendation. The proposed scheme can calculate the similarity between users to determine their potential relationships and interests, and it also can protect the users' privacy from leaking to an untrusted third party. The security analysis and experimental results showed that our proposed scheme provides excellent performance and is feasible for real-world applications.

cs.CR↗

arXiv2025

X-WebAgentBench: A Multilingual Interactive Web Benchmark for Evaluating Global Agentic System

Recently, large language model (LLM)-based agents have achieved significant success in interactive environments, attracting significant academic and industrial attention. Despite these advancements, current research predominantly focuses on English scenarios. In reality, there are over 7,000 languages worldwide, all of which demand access to comparable agentic services. Nevertheless, the development of language agents remains inadequate for meeting the diverse requirements of multilingual agentic applications. To fill this gap, we introduce X-WebAgentBench, a novel multilingual agent benchmark in an interactive web environment, which evaluates the planning and interaction performance of language agents across multiple languages, thereby contributing to the advancement of global agent intelligence. Additionally, we assess the performance of various LLMs and cross-lingual alignment methods, examining their effectiveness in enhancing agents. Our findings reveal that even advanced models like GPT-4o, when combined with cross-lingual techniques, fail to achieve satisfactory results. We hope that X-WebAgentBench can serve as a valuable benchmark for multilingual agent scenario in real-world applications.

cs.CL↗

arXiv2025

Demonstrating Coherent Quantum Routers for Bucket-Brigade Quantum Random Access Memory on a Superconducting Processor

Quantum routers (QRouters) are essential components of bucket-brigade quantum random access memory (QRAM), enabling quantum applications such as Grover's search and quantum machine learning. Despite significant theoretical advances, achieving scalable and coherent QRouters experimentally remains challenging. Here, we demonstrate coherent quantum routers using a superconducting quantum processor, laying a practical foundation for scalable QRAM systems. The quantum router at the core of our implementation utilizes the transition composite gate (TCG) scheme, wherein auxiliary energy levels temporarily mediate conditional interactions, substantially reducing circuit depth compared to traditional gate decompositions. Moreover, by encoding routing addresses in the non-adjacent qutrit states $|0\rangle$ and $|2\rangle$, our design inherently enables eraser-detection capability, providing efficient post-selection to mitigate routing errors. Experimentally, we achieve individual QRouter fidelities up to 95.74%, and validate scalability through a two-layer quantum routing network achieving an average fidelity of 82.40%. Our results represent a significant advancement in quantum routing technology, providing enhanced fidelity, built-in error resilience, and practical scalability crucial for the development of future QRAM and large-scale quantum computing architectures.

quant-ph↗

arXiv2025

LLM A: Human in the Loop Large Language Models Enabled A Search for Robotics

This research focuses on how Large Language Models (LLMs) can help with (path) planning for mobile embodied agents such as robots, in a human-in-the-loop and interactive manner. A novel framework named LLM A*, aims to leverage the commonsense of LLMs, and the utility-optimal A* is proposed to facilitate few-shot near-optimal path planning. Prompts are used for two main purposes: 1) to provide LLMs with essential information like environments, costs, heuristics, etc.; 2) to communicate human feedback on intermediate planning results to LLMs. This approach takes human feedback on board and renders the entire planning process transparent (akin to a `white box') to humans. Moreover, it facilitates code-free path planning, thereby fostering the accessibility and inclusiveness of artificial intelligence techniques to communities less proficient in coding. Comparative analysis against A* and RL demonstrates that LLM A* exhibits greater efficiency in terms of search space and achieves paths comparable to A* while outperforming RL. The interactive nature of LLM A* also makes it a promising tool for deployment in collaborative human-robot tasks. Codes and Supplemental Materials can be found at GitHub: https://github.com/speedhawk/LLM-A-.

cs.RO↗

arXiv2025

Qwen3 Technical Report

In this work, we present Qwen3, the latest version of the Qwen model family. Qwen3 comprises a series of large language models (LLMs) designed to advance performance, efficiency, and multilingual capabilities. The Qwen3 series includes models of both dense and Mixture-of-Expert (MoE) architectures, with parameter scales ranging from 0.6 to 235 billion. A key innovation in Qwen3 is the integration of thinking mode (for complex, multi-step reasoning) and non-thinking mode (for rapid, context-driven responses) into a unified framework. This eliminates the need to switch between different models--such as chat-optimized models (e.g., GPT-4o) and dedicated reasoning models (e.g., QwQ-32B)--and enables dynamic mode switching based on user queries or chat templates. Meanwhile, Qwen3 introduces a thinking budget mechanism, allowing users to allocate computational resources adaptively during inference, thereby balancing latency and performance based on task complexity. Moreover, by leveraging the knowledge from the flagship models, we significantly reduce the computational resources required to build smaller-scale models, while ensuring their highly competitive performance. Empirical evaluations demonstrate that Qwen3 achieves state-of-the-art results across diverse benchmarks, including tasks in code generation, mathematical reasoning, agent tasks, etc., competitive against larger MoE models and proprietary models. Compared to its predecessor Qwen2.5, Qwen3 expands multilingual support from 29 to 119 languages and dialects, enhancing global accessibility through improved cross-lingual understanding and generation capabilities. To facilitate reproducibility and community-driven research and development, all Qwen3 models are publicly accessible under Apache 2.0.

cs.CL↗

arXiv2025

PRIME: Physics-Related Intelligent Mixture of Experts for Transistor Characteristics Prediction

In recent years, machine learning has been extensively applied to data prediction during process ramp-up, with a particular focus on transistor characteristics for circuit design and manufacture. However, capturing the nonlinear current response across multiple operating regions remains a challenge for neural networks. To address such challenge, a novel machine learning framework, PRIME (Physics-Related Intelligent Mixture of Experts), is proposed to capture and integrate complex regional characteristics. In essence, our framework incorporates physics-based knowledge with data-driven intelligence. By leveraging a dynamic weighting mechanism in its gating network, PRIME adaptively activates the suitable expert model based on distinct input data features. Extensive evaluations are conducted on various gate-all-around (GAA) structures to examine the effectiveness of PRIME and considerable improvements (60\%-84\%) in prediction accuracy are shown over state-of-the-art models.

cs.LG↗

arXiv2025

Cosmic filament spin -- I: a comparative study in observation

In the cosmic web, filaments play a crucial role in connecting walls to clusters and also act as an important stage for galaxy formation and evolution. Recent observational studies claim that filaments have spin. In this study, we examined the potential impact of diversity in filament identification algorithms and galaxy survey datasets on the quantification of filament spin. The results of this study demonstrate qualitative agreement with previous research, suggesting that a reliable filament spin signal is detectable when the viewing angle of filament spine larger than 80 degrees under a rough estimation. The detected filament spin signal is intricately linked to the viewing angle, dynamic temperature, etc. The quantitative difference of filament spin signal among samples is slightly dependent on the filament identification algorithms, while the value is relatively greater dependent on the redshift space distortion effect in the galaxy sample.

astro-ph.CO↗

arXiv2025

Cosmic filament spin -- II: filament spin and its impact on galaxy spin-filament alignment in a cosmological simulation

Observational studies have reported that cosmic filaments on the megaparsec scale exhibit rotational motion. Subsequent simulation studies have shown qualitative agreement with these findings, but quantitative discrepancies remain due to differences in data and methods, which require verification. To address this issue, we adopt the same methodology as used in the observations to identify filament spin from the galaxy distribution constructed from a hydrodynamic simulation. Using the same approach to measure filament spin, we find that the simulation results closely match the observational findings, with only minor discrepancies arising from slight differences in the fraction of filaments classified as dynamically cold or hot based on their dynamic temperature. Additionally, an analysis of how filament spin affects the galaxy spin-filament correlation shows that filaments with strong spin signals and dynamically cold have a greater impact on the galaxy spin-filament correlation than those with weaker spin signals and dynamically hot filaments. These results not only provide further evidence that cosmic filaments exhibit spin, but also highlight the importance of this rotation in the acquisition of angular momentum by individual galaxies. Future studies exploring the influence of filament spin on galaxy spin may shed light on the physical origins of filaments and the angular momentum of galaxies.

astro-ph.CO↗

arXiv2025

Drug classification based on X-ray spectroscopy combined with machine learning

The proliferation of new types of drugs necessitates the urgent development of faster and more accurate detection methods. Traditional detection methods have high requirements for instruments and environments, making the operation complex. X-ray absorption spectroscopy, a non-destructive detection technique, offers advantages such as ease of operation, penetrative observation, and strong substance differentiation capabilities, making it well-suited for application in the field of drug detection and identification. In this study, we constructed a classification model using Convolutional Neural Networks (CNN), Support Vector Machines (SVM), and Particle Swarm Optimization (PSO) to classify and identify drugs based on their X-ray spectral profiles. In the experiments, we selected 14 chemical reagents with chemical formulas similar to drugs as samples. We utilized CNN to extract features from the spectral data of these 14 chemical reagents and used the extracted features to train an SVM model. We also utilized PSO to optimize two critical initial parameters of the SVM. The experimental results demonstrate that this model achieved higher classification accuracy compared to two other common methods, with a prediction accuracy of 99.14%. Additionally, the model exhibited fast execution speed, mitigating the drawback of a drastic increase in running time and efficiency reduction that may result from the direct fusion of PSO and SVM. Therefore, the combined approach of X-ray absorption spectroscopy with CNN, PSO, and SVM provides a rapid, highly accurate, and reliable classification and identification method for the field of drug detection, holding promising prospects for widespread application.

cs.CV↗