arXiv subjects

Jun Wang

Publications and source records attributed to Jun Wang.

At least 235 records · Page 13Linked to original sources

arXiv2026

Combined analysis of the singly-Cabbibo-suppressed decays of $D^{0} \to VP$

We investigate six singly Cabibbo-suppressed decay channels in $D^0\to VP$ ( $V$ and $P$ stand for the ground state vector and pseudoscalar mesons, respectively), i.e. $D^{0}\to ρ^{+}π^{-}$, $ρ^{-}π^{+}$, $K^{*+}K^{-}$, $K^{*-}K^{+}$, $K^{*0}\bar{K}^{0}$, and $\bar{K}^{*0}K^{0}$. These decay channels share the similar transition mechanisms involving only the direct emission (DE) and internal conversion (IC) processes. We show that a combined analysis of these channels can explicitly highlight the role played by the IC processes which contribute to the amplitudes at the same order of magnitude as the DE processes.

hep-ph↗

arXiv2026

Frequency Error-Guided Under-sampling Optimization for Multi-Contrast MRI Reconstruction

Magnetic resonance imaging (MRI) plays a vital role in clinical diagnostics, yet it remains hindered by long acquisition times and motion artifacts. Multi-contrast MRI reconstruction has emerged as a promising direction by leveraging complementary information from fully-sampled reference scans. However, existing approaches suffer from three major limitations: (1) superficial reference fusion strategies, such as simple concatenation, (2) insufficient utilization of the complementary information provided by the reference contrast, and (3) fixed under-sampling patterns. We propose an efficient and interpretable frequency error-guided reconstruction framework to tackle these issues. We first employ a conditional diffusion model to learn a Frequency Error Prior (FEP), which is then incorporated into a unified framework for jointly optimizing both the under-sampling pattern and the reconstruction network. The proposed reconstruction model employs a model-driven deep unfolding framework that jointly exploits frequency- and image-domain information. In addition, a spatial alignment module and a reference feature decomposition strategy are incorporated to improve reconstruction quality and bridge model-based optimization with data-driven learning for improved physical interpretability. Comprehensive validation across multiple imaging modalities, acceleration rates (4-30x), and sampling schemes demonstrates consistent superiority over state-of-the-art methods in both quantitative metrics and visual quality. All codes are available at https://github.com/fangxinming/JUF-MRI.

cs.CV↗

arXiv2026

Apollo: Unified Multi-Task Audio-Video Joint Generation

Audio-video joint generation has progressed rapidly, yet substantial challenges still remain. Non-commercial approaches still suffer audio-visual asynchrony, poor lip-speech alignment, and unimodal degradation, which can be stemmed from weak audio-visual correspondence modeling, limited generalization, and scarce high-quality dense-caption data. To address these issues, we introduce Apollo and delve into three axes--model architecture, training strategy, and data curation. Architecturally, we adopt a single-tower design with unified DiT blocks and an Omni-Full Attention mechanism, achieving tight audio-visual alignment and strong scalability. Training-wise, we adopt a progressive multitask regime--random modality masking to joint optimization across tasks, and a multistage curriculum, yielding robust representations, strengthening A-V aligned world knowledge, and preventing unimodal collapse. For datasets, we present the first large-scale audio-video dataset with dense captions, and introduce a novel automated data-construction pipeline which annotates and filters millions of diverse, high-quality, strictly aligned audio-video-caption triplets. Building on this, Apollo scales to large datasets, delivering high-fidelity, semantically and temporally aligned, instruction-following generation in both joint and unimodal settings while generalizing robustly to out-of-distribution scenarios. Across tasks, it substantially outperforms prior methods by a large margin and achieves performance comparable to Veo 3, offering a unified, scalable path toward next-generation audio-video synthesis.

cs.CV↗

arXiv2026

Markovian Pre-Trained Transformer for Next-Item Recommendation

We introduce the Markovian Pre-trained Transformer (MPT) for next-item recommendation, a transferable model fully pre-trained on synthetic Markov chains, yet capable of achieving state-of-the-art performance by fine-tuning a lightweight adaptor. This counterintuitive success stems from the observation of the `Markovian' nature: advanced sequential recommenders coincidentally rely on the latest interaction to make predictions, while the historical interactions serve mainly as auxiliary cues for inferring the user's general, non-sequential identity. This characteristic necessitates the capabilities of a universal recommendation model to effectively summarize the user sequence, with particular emphasis on the latest interaction. MPT inherently has the potential to be universal and transferable. On the one hand, when trained to predict the next state of Markov chains, it acquires the capabilities to estimate transition probabilities from the context (one adaptive manner for summarizing sequences) and attend to the last state to ensure accurate state transitions. On the other hand, unlike the heterogeneous interaction data, an unlimited amount of controllable Markov chains is available to boost the model capacity. We conduct extensive experiments on five public datasets from three distinct platforms to validate the superiority of Markovian pre-training over traditional recommendation pre-training and recent language pre-training paradigms.

cs.IR↗

arXiv2026

A Brain-like Synergistic Core in LLMs Drives Behaviour and Learning

The independent evolution of intelligence in biological and artificial systems offers a unique opportunity to identify its fundamental computational principles. Here we show that large language models spontaneously develop synergistic cores -- components where information integration exceeds individual parts -- remarkably similar to those in the human brain. Using principles of information decomposition across multiple LLM model families and architectures, we find that areas in middle layers exhibit synergistic processing while early and late layers rely on redundancy, mirroring the informational organisation in biological brains. This organisation emerges through learning and is absent in randomly initialised networks. Crucially, ablating synergistic components causes disproportionate behavioural changes and performance loss, aligning with theoretical predictions about the fragility of synergy. Moreover, fine-tuning synergistic regions through reinforcement learning yields significantly greater performance gains than training redundant components, yet supervised fine-tuning shows no such advantage. This convergence suggests that synergistic information processing is a fundamental property of intelligence, providing targets for principled model design and testable predictions for biological intelligence.

cs.AI↗

arXiv2026

Knowledge-Driven Multi-Turn Jailbreaking on Large Language Models

Large Language Models (LLMs) face a significant threat from multi-turn jailbreak attacks, where adversaries progressively steer conversations to elicit harmful outputs. However, the practical effectiveness of existing attacks is undermined by several critical limitations: they struggle to maintain a coherent progression over long interactions, often losing track of what has been accomplished and what remains to be done; they rely on rigid or pre-defined patterns, and fail to adapt to the LLM's dynamic and unpredictable conversational state. To address these shortcomings, we introduce Mastermind, a multi-turn jailbreak framework that adopts a dynamic and self-improving approach. Mastermind operates in a closed loop of planning, execution, and reflection, enabling it to autonomously build and refine its knowledge of model vulnerabilities through interaction. It employs a hierarchical planning architecture that decouples high-level attack objectives from low-level tactical execution, ensuring long-term focus and coherence. This planning is guided by a knowledge repository that autonomously discovers and refines effective attack patterns by reflecting on interactive experiences. Mastermind leverages this accumulated knowledge to dynamically recombine and adapt attack vectors, dramatically improving both effectiveness and resilience. We conduct comprehensive experiments against state-of-the-art models, including GPT-5 and Claude 3.7 Sonnet. The results demonstrate that Mastermind significantly outperforms existing baselines, achieving substantially higher attack success rates and harmfulness ratings. Moreover, our framework exhibits notable resilience against multiple advanced defense mechanisms.

cs.CR↗

arXiv2026

MM-Sonate: Multimodal Controllable Audio-Video Generation with Zero-Shot Voice Cloning

Joint audio-video generation aims to synthesize synchronized multisensory content, yet current unified models struggle with fine-grained acoustic control, particularly for identity-preserving speech. Existing approaches either suffer from temporal misalignment due to cascaded generation or lack the capability to perform zero-shot voice cloning within a joint synthesis framework. In this work, we present MM-Sonate, a multimodal flow-matching framework that unifies controllable audio-video joint generation with zero-shot voice cloning capabilities. Unlike prior works that rely on coarse semantic descriptions, MM-Sonate utilizes a unified instruction-phoneme input to enforce strict linguistic and temporal alignment. To enable zero-shot voice cloning, we introduce a timbre injection mechanism that effectively decouples speaker identity from linguistic content. Furthermore, addressing the limitations of standard classifier-free guidance in multimodal settings, we propose a noise-based negative conditioning strategy that utilizes natural noise priors to significantly enhance acoustic fidelity. Empirical evaluations demonstrate that MM-Sonate establishes new state-of-the-art performance in joint generation benchmarks, significantly outperforming baselines in lip synchronization and speech intelligibility, while achieving voice cloning fidelity comparable to specialized Text-to-Speech systems.

cs.SD↗

arXiv2026

Spider web-inspired sensing and computation with fiber network physical reservoirs

Physical reservoir computing leverages the intrinsic dynamics of mechanical systems to perform computation through their natural responses to input signals. Here, we study a compliant fiber network inspired by orb-weaving spider webs and investigate how its mechanical design and operating conditions shape its computational capability. Using Cosserat rod-based simulations, we identify how network topology, geometry, actuation, and axial tension impact the nonlinear computation and memory capacity of the network. We further evaluate several readout reduction strategies to assess how computational performance varies with the number and placement of measured outputs. We then experimentally validate these results using a physical fiber-network prototype. Overall, results provide insights and guidance on design, actuation, and sensing choices to enable fiber networks for mechano-intelligent computation. They demonstrate the ability of structured compliant fibers networks to serve as physical reservoirs capable of nonlinear transformation and input-history retention.

eess.SY↗

arXiv2026

OThink-R1: Intrinsic Fast/Slow Thinking Mode Switching for Over-Reasoning Mitigation

Human cognition operates through two complementary modes: fast intuitive thinking and slow deliberate thinking. Vanilla large language models (LLMs) predominantly follow the fast-thinking paradigm, producing immediate responses; while recent large reasoning models (LRMs) adopt slow-thinking strategies, generating detailed reasoning chains before arriving at answers. While LRMs often achieve higher accuracy, this comes at the cost of substantially increased token usage. To address this efficiency-accuracy trade-off, we propose OThink-R1, a hybrid reasoning framework that integrates both modes within a single LRM and enables automatic mode switching based on problem characteristics. We first identify three major patterns of essential and redundant reasoning trajectories in LRMs, which guide the design of an auxiliary LLM-based judge that adaptively determines when slow thinking is necessary. Leveraging the judge's decisions, we construct a hybrid fine-tuning dataset by pruning redundant reasoning to produce fast-thinking samples and retaining complete reasoning for slow-thinking samples. This dataset is then used to fine-tune LRMs, equipping them with inherent autonomous mode-selection capabilities. Extensive experiments on mathematical and question-answering benchmarks show that OThink-R1 reduces reasoning token usage significantly while maintaining competitive accuracy. The code is available at https://github.com/AgenticIR-Lab/OThink-R1.

cs.AI↗

arXiv2026

Vision-Language Reasoning for Geolocalization: A Reinforcement Learning Approach

Recent advances in vision-language models have opened up new possibilities for reasoning-driven image geolocalization. However, existing approaches often rely on synthetic reasoning annotations or external image retrieval, which can limit interpretability and generalizability. In this paper, we present Geo-R, a retrieval-free framework that uncovers structured reasoning paths from existing ground-truth coordinates and optimizes geolocation accuracy via reinforcement learning. We propose the Chain of Region, a rule-based hierarchical reasoning paradigm that generates precise, interpretable supervision by mapping GPS coordinates to geographic entities (e.g., country, province, city) without relying on model-generated or synthetic labels. Building on this, we introduce a lightweight reinforcement learning strategy with coordinate-aligned rewards based on Haversine distance, enabling the model to refine predictions through spatially meaningful feedback. Our approach bridges structured geographic reasoning with direct spatial supervision, yielding improved localization accuracy, stronger generalization, and more transparent inference. Experimental results across multiple benchmarks confirm the effectiveness of Geo-R, establishing a new retrieval-free paradigm for scalable and interpretable image geolocalization. To facilitate further research and ensure reproducibility, both the model and code will be made publicly available.

cs.CL↗

arXiv2026

InfMasking: Unleashing Synergistic Information by Contrastive Multimodal Interactions

In multimodal representation learning, synergistic interactions between modalities not only provide complementary information but also create unique outcomes through specific interaction patterns that no single modality could achieve alone. Existing methods may struggle to effectively capture the full spectrum of synergistic information, leading to suboptimal performance in tasks where such interactions are critical. This is particularly problematic because synergistic information constitutes the fundamental value proposition of multimodal representation. To address this challenge, we introduce InfMasking, a contrastive synergistic information extraction method designed to enhance synergistic information through an Infinite Masking strategy. InfMasking stochastically occludes most features from each modality during fusion, preserving only partial information to create representations with varied synergistic patterns. Unmasked fused representations are then aligned with masked ones through mutual information maximization to encode comprehensive synergistic information. This infinite masking strategy enables capturing richer interactions by exposing the model to diverse partial modality combinations during training. As computing mutual information estimates with infinite masking is computationally prohibitive, we derive an InfMasking loss to approximate this calculation. Through controlled experiments, we demonstrate that InfMasking effectively enhances synergistic information between modalities. In evaluations on large-scale real-world datasets, InfMasking achieves state-of-the-art performance across seven benchmarks. Code is released at https://github.com/brightest66/InfMasking.

cs.LG↗

arXiv2026

$I=\frac{3}{2}$ $πK$ $s$-wave scattering length from lattice QCD

The $I=\frac{3}{2}$ $πK$ $s$-wave scattering phase shift is computed by lattice quantum chromodynamics with $N_f=3$ flavors of Asqtad-improved staggered fermions. The energy-eigenvalues of $πK$ systems at one center of mass frame and six moving frames using moving wall source technique are used to get phase shifts by Lüscher's formula and its extensions. The calculations are good enough to acquire effective range expansion parameters: scattering length $a$, effective range $r$, and shape parameter $P$, which are in good agreement with our explicit analytical predictions in three-flavor chiral perturbation theory at next-to-leading order. All results are fairly consistent with experimental measurements, phenomenological studies, and lattice estimations. Numerical computations are implemented at a fine ($a\approx0.082$ fm, $L^3 T = 40^3 96$) lattice ensemble with physical quark masses.

hep-lat↗

arXiv2025

Constrained Language Model Policy Optimization via Risk-aware Stepwise Alignment

When fine-tuning pre-trained Language Models (LMs) to exhibit desired behaviors, maintaining control over risk is critical for ensuring both safety and trustworthiness. Most existing safety alignment methods, such as Safe RLHF and SACPO, typically operate under a risk-neutral paradigm that is insufficient to address the risks arising from deviations from the reference policy and offers limited robustness against rare but potentially catastrophic harmful behaviors. To address this limitation, we propose Risk-aware Stepwise Alignment (RSA), a novel alignment method that explicitly incorporates risk awareness into the policy optimization process by leveraging a class of nested risk measures. Specifically, RSA formulates safety alignment as a token-level risk-aware constrained policy optimization problem and solves it through a stepwise alignment procedure that yields token-level policy updates derived from the nested risk measures. This design offers two key benefits: (1) it mitigates risks induced by excessive model shift away from a reference policy, and (2) it explicitly suppresses low-probability yet high-impact harmful behaviors. Moreover, we provide theoretical analysis on policy optimality under mild assumptions. Experimental results demonstrate that our method achieves high levels of helpfulness while ensuring strong safety and significantly suppresses tail risks, namely low-probability yet high-impact unsafe responses.

cs.AI↗

arXiv2025

Electromagnetically-Induced Transparency Bridges Disconnected Light-Harvesting Networks

The energy-transfer efficiency of the natural photosynthesis system seems to be perfectly optimized during the evolution for millions of years. However, how to enhance the efficiency in the artificial light-harvesting systems is still unclear. In this paper, we investigate the energy-transfer process in the photosystem I (PSI). When there is no effective coupling between the outer antenna (OA) and the reaction center (RC), the two light-harvesting networks are disconnected and thus the energy transfer is inefficient. In order to repair these disconnected networks, we introduce a bridge with three sites between them. We find that by modulating the level structure of the 3-site bridge to be resonant, the energy transfer via the dark state will be enhanced and even outperform the original PSI. Our discoveries may shed light on the designing mechanism of artificial light-harvesting systems.

physics.optics↗

arXiv2025

A new iterated Tikhonov regularization method for Fredholm integral equation of first kind

We consider Fredholm integral equation of the first kind, present an efficient new iterated Tikhonov method to solve it. The new Tikhonov iteration method has been proved which can achieve the optimal order under a-priori assumption. In numerical experiments, the new iterated Tikhonov regularization method is compared with the classical iterated Tikhonov method, Landweber iteration method to solve the corresponding discrete problem, which indicates the validity and efficiency of the proposed method.

math.NA↗

arXiv2025

Bridging the Copyright Gap: Do Large Vision-Language Models Recognize and Respect Copyrighted Content?

Large vision-language models (LVLMs) have achieved remarkable advancements in multimodal reasoning tasks. However, their widespread accessibility raises critical concerns about potential copyright infringement. Will LVLMs accurately recognize and comply with copyright regulations when encountering copyrighted content (i.e., user input, retrieved documents) in the context? Failure to comply with copyright regulations may lead to serious legal and ethical consequences, particularly when LVLMs generate responses based on copyrighted materials (e.g., retrieved book experts, news reports). In this paper, we present a comprehensive evaluation of various LVLMs, examining how they handle copyrighted content -- such as book excerpts, news articles, music lyrics, and code documentation when they are presented as visual inputs. To systematically measure copyright compliance, we introduce a large-scale benchmark dataset comprising 50,000 multimodal query-content pairs designed to evaluate how effectively LVLMs handle queries that could lead to copyright infringement. Given that real-world copyrighted content may or may not include a copyright notice, the dataset includes query-content pairs in two distinct scenarios: with and without a copyright notice. For the former, we extensively cover four types of copyright notices to account for different cases. Our evaluation reveals that even state-of-the-art closed-source LVLMs exhibit significant deficiencies in recognizing and respecting the copyrighted content, even when presented with the copyright notice. To solve this limitation, we introduce a novel tool-augmented defense framework for copyright compliance, which reduces infringement risks in all scenarios. Our findings underscore the importance of developing copyright-aware LVLMs to ensure the responsible and lawful use of copyrighted content.

cs.CL↗

arXiv2025

PCART: Automated Repair of Python API Parameter Compatibility Issues

In modern software development, Python third-party libraries play a critical role, especially in fields like deep learning and scientific computing. However, API parameters in these libraries often change during evolution, leading to compatibility issues for client applications reliant on specific versions. Python's flexible parameter-passing mechanism further complicates this, as different passing methods can result in different API compatibility. Currently, no tool can automatically detect and repair Python API parameter compatibility issues. To fill this gap, we introduce PCART, the first solution to fully automate the process of API extraction, code instrumentation, API mapping establishment, compatibility assessment, repair, and validation. PCART handles various types of Python API parameter compatibility issues, including parameter addition, removal, renaming, reordering, and the conversion of positional to keyword parameters. To evaluate PCART, we construct PCBENCH, a large-scale benchmark comprising 47,478 test cases mutated from 844 parameter-changed APIs across 33 popular Python libraries. Evaluation results demonstrate that PCART is both effective and efficient, significantly outperforming existing tools (MLCatchUp and Relancer) and the large language model ChatGPT (GPT-4o), achieving an F1-score of 96.51% in detecting API parameter compatibility issues and a repair precision of 91.97%. Further evaluation on 30 real-world Python projects from GitHub confirms PCART's practicality. We believe PCART can significantly reduce the time programmers spend maintaining Python API updates and advance the automation of Python API compatibility issue repair.

cs.SE↗

arXiv2025

Adaptive Command: Real-Time Policy Adjustment via Language Models in StarCraft II

We present Adaptive Command, a novel framework integrating large language models (LLMs) with behavior trees for real-time strategic decision-making in StarCraft II. Our system focuses on enhancing human-AI collaboration in complex, dynamic environments through natural language interactions. The framework comprises: (1) an LLM-based strategic advisor, (2) a behavior tree for action execution, and (3) a natural language interface with speech capabilities. User studies demonstrate significant improvements in player decision-making and strategic adaptability, particularly benefiting novice players and those with disabilities. This work contributes to the field of real-time human-AI collaborative decision-making, offering insights applicable beyond RTS games to various complex decision-making scenarios.

cs.HC↗