arXiv subjects

Qiang Zhou

Publications and source records attributed to Qiang Zhou.

At least 37 records · Page 2Linked to original sources

arXiv2026

MigrationBench: Repository-Level Code Migration Benchmark from Java 8

With the rapid advancement of powerful large language models (LLMs) in recent years, a wide range of software engineering tasks can now be addressed using LLMs, significantly enhancing productivity and scalability. Numerous benchmark datasets have been developed to evaluate the coding capabilities of these models, while they primarily focus on code generation and issue-resolution tasks. In contrast, we introduce a new coding benchmark MigrationBench with a distinct focus: code migration. MigrationBench aims to serve as a comprehensive benchmark for migration from Java 8 to the latest long-term support (LTS) versions (Java 17, 21), including a full dataset and its subset selected with 5,102 and 300 repositories respectively. selected is a representative subset curated for complexity and difficulty, offering a versatile resource to support research in the field of code migration. Additionally, we provide a comprehensive evaluation framework to facilitate rigorous and standardized assessment of LLMs on this challenging task. We further propose an agentic framework and demonstrate that LLMs can effectively tackle repository-level code migration to Java 17. For the selected subset with Claude-4.5-Sonnet, our agentic framework achieves 71.67% and 53.33% success rate (pass@1) for minimal and maximal migration respectively. The dataset and evaluation source code are available at: https://huggingface.co/collections/AmazonScience/migrationbench and https://github.com/amazon-science/MigrationBench respectively.

cs.SE↗

arXiv2026

Integrated time-bin entangled quantum light source on a 4H-SiC microring chip

Integrated time-bin-entangled photon-pair source with cavity-enhanced nonlinear optical processes is essential for quantum information technologies. However, microcavities with a high quality factor inherently introduce a trade-off between generation efficiency and photon bandwidth, which hinders the development of high-speed quantum networks with an integrated source. Here, we address this challenge by optimizing the nonlinearity property of the material and the geometry of the integrated microring resonator with a 4H-silicon carbide platform. Operating at a loaded quality factor of 1.9 $\times$ 10^5 - spectral bandwidth of 1.0 GHz and pumped with 300-ps double pulses separated by 1.25 ns at a repetition rate of 160 MHz, the device achieves a time-bin-entangled photon-pair generation rate of 1.35 $\times$ 10^7 s^-1 mW^-2. A raw visibility of 95.55 $\pm$ 0.18% is measured, showing a violation of Bell's inequality by more than 138 standard deviations, and a fidelity of 94.37 $\pm$ 0.22% is obtained by quantum state tomography. These results provide a scalable pathway to an efficient and broadband time-bin entangled quantum light source, overcoming intrinsic limitations of cavity-based designs and advancing integrated platforms for future quantum communication networks.

quant-ph↗

arXiv2026

DDA-Thinker: Decoupled Dual-Atomic Reinforcement Learning for Reasoning-Driven Image Editing

Recent image editing models have achieved strong visual fidelity but often struggle with tasks requiring complex reasoning. To investigate and enhance the reasoning-grounded planning for image editing, we propose DDA-Thinker, a Thinker-centric framework designed for the independent optimization of a planning module (Thinker) over a fixed generative model (Editor). This decoupled Thinker-centric paradigm facilitates a controlled analysis of the planning module and makes its contribution under a fixed Editor easier to assess. To effectively guide this Thinker, we introduce a dual-atomic reinforcement learning framework. This framework decomposes feedback into two distinct atomic rewards implemented through verifiable checklists: a cognitive-atomic reward to directly assess the quality of the Thinker's executable plan, which serves as the actionable outcome of the Thinker's reasoning, and a visual-atomic reward to assess the final image quality. To improve checklist quality, our checklist synthesis is grounded not only in the source image and user instruction but also in a rational reference description of the ideal post-edit scene. To support this training, we further develop a two-stage data curation pipeline that first synthesizes a diverse and reasoning-focused dataset, then applies difficulty-aware refinement to curate an effective training curriculum for reinforcement learning. Extensive experiments on reasoning-driven image editing benchmarks, including RISE-Bench and KRIS-Bench, demonstrate that our approach substantially improves overall performance. Our method enables a community model to achieve results competitive with strong proprietary models, highlighting the practical potential of Thinker-centric optimization under a fixed-editor setting.

cs.CV↗

arXiv2026

Change-of-Rings Theorems for the Small Finitistic Dimension

In this paper, we study the small finitistic dimension of a commutative ring from the viewpoint of finitistic flat homological algebra. Using the class $FPR(R)$ of modules admitting finite projective resolutions, we investigate the finitistic flat ($FT$-flat) dimension and establish several of its basic properties. We prove change-of-rings results for the $FT$-flat dimension, including quotient and polynomial extension results, as well as localization inequalities. As applications, we obtain characterizations of the small finitistic dimension in terms of $FT$-flat dimension, derive quotient and polynomial extension theorems for the small finitistic dimension, and establish local upper bounds in terms of the small finitistic dimensions of localizations.

math.AC↗

arXiv2026

Unified Thinker: A General Reasoning Modular Core for Image Generation

Despite impressive progress in high-fidelity image synthesis, generative models still struggle with logic-intensive instruction following, exposing a persistent reasoning--execution gap. Meanwhile, closed-source systems (e.g., Nano Banana) have demonstrated strong reasoning-driven image generation, highlighting a substantial gap to current open-source models. We argue that closing this gap requires not merely better visual generators, but executable reasoning: decomposing high-level intents into grounded, verifiable plans that directly steer the generative process. To this end, we propose Unified Thinker, a task-agnostic reasoning architecture for general image generation, designed as a unified planning core that can plug into diverse generators and workflows. Unified Thinker decouples a dedicated Thinker from the image Generator, enabling modular upgrades of reasoning without retraining the entire generative model. We further introduce a two-stage training paradigm: we first build a structured planning interface for the Thinker, then apply reinforcement learning to ground its policy in pixel-level feedback, encouraging plans that optimize visual correctness over textual plausibility. Extensive experiments on text-to-image generation and image editing show that Unified Thinker substantially improves image reasoning and generation quality.

cs.CV↗

arXiv2026

SpatialReward: Verifiable Spatial Reward Modeling for Fine-Grained Spatial Consistency in Text-to-Image Generation

Recent advances in text-to-image (T2I) generation via reinforcement learning (RL) have benefited from reward models that assess semantic alignment and visual quality. However, most existing reward models pay limited attention to fine-grained spatial relationships, often producing images that appear plausible overall yet contain inaccuracies in object positioning. In this work, we present \textbf{SpatialReward}, a verifiable reward model explicitly designed to evaluate spatial layouts in generated images. SpatialReward adopts a multi-stage pipeline: a \emph{Prompt Decomposer} extracts entities, attributes, and spatial metadata from free-form prompts; expert detectors provide accurate visual grounding of object positions and attributes; and a vision-language model applies chain-of-thought reasoning over grounded observations to assess complex spatial relations that are challenging for rule-based methods. To more comprehensively evaluate spatial relationships in generated images, we introduce \textbf{SpatRelBench}, a benchmark covering object attributes, orientation, inter-object relations, and rendered text placement. Experiments on Stable Diffusion and FLUX show that incorporating SpatialReward into RL training consistently improves spatial consistency and overall generation quality, with results aligned more closely to human judgments. These findings indicate that verifiable reward models hold considerable potential for enabling more accurate and controllable optimization in text-to-image generation models.

cs.CV↗

arXiv2026

Contribution-aware Token Compression for Efficient Video Understanding via Reinforcement Learning

Video large language models have demonstrated remarkable capabilities in video understanding tasks. However, the redundancy of video tokens introduces significant computational overhead during inference, limiting their practical deployment. Many compression algorithms are proposed to prioritize retaining features with the highest attention scores to minimize perturbations in attention computations. However, the correlation between attention scores and their actual contribution to correct answers remains ambiguous. To address the above limitation, we propose a novel \textbf{C}ontribution-\textbf{a}ware token \textbf{Co}mpression algorithm for \textbf{VID}eo understanding (\textbf{CaCoVID}) that explicitly optimizes the token selection policy based on the contribution of tokens to correct predictions. First, we introduce a reinforcement learning-based framework that optimizes a policy network to select video token combinations with the greatest contribution to correct predictions. This paradigm shifts the focus from passive token preservation to active discovery of optimal compressed token combinations. Secondly, we propose a combinatorial policy optimization algorithm with online combination space sampling, which dramatically reduces the exploration space for video token combinations and accelerates the convergence speed of policy optimization. Extensive experiments on diverse video understanding benchmarks demonstrate the effectiveness of CaCoVID. Codes are available at https://github.com/LivingFutureLab/CaCoVID.

cs.CV↗

arXiv2025

SEEA-R1: Tree-Structured Reinforcement Fine-Tuning for Self-Evolving Embodied Agents

Self-evolution, the ability of agents to autonomously improve their reasoning and behavior, is essential for the embodied domain with long-horizon, real-world tasks. Despite current advancements in reinforcement fine-tuning (RFT) showing strong performance in enhancing reasoning in LLMs, its potential to enable self-evolving embodied intelligence with multi-modal interactions remains largely unexplored. Specifically, reinforcement fine-tuning faces two fundamental obstacles in embodied settings: (i) the lack of accessible intermediate rewards in multi-step reasoning tasks limits effective learning signals, and (ii) reliance on hand-crafted reward functions restricts generalization to novel tasks and environments. To address these challenges, we present Self-Evolving Embodied Agents-R1, SEEA-R1, the first RFT framework designed for enabling the self-evolving capabilities of embodied agents. Specifically, to convert sparse delayed rewards into denser intermediate signals that improve multi-step reasoning, we propose Tree-based group relative policy optimization (Tree-GRPO) integrates Monte Carlo Tree Search into GRPO. To generalize reward estimation across tasks and scenes, supporting autonomous adaptation and reward-driven self-evolution, we further introduce Multi-modal Generative Reward Model (MGRM). To holistically evaluate the effectiveness of SEEA-R1, we evaluate on the ALFWorld benchmark, surpassing state-of-the-art methods with scores of 85.07% (textual) and 46.27% (multi-modal), outperforming prior models including GPT-4o. SEEA-R1 also achieves scores of 80.3% (textual) and 44.03% (multi-modal) without ground truth reward, surpassing all open-source baselines and highlighting its scalability as a self-evolving embodied agent. Additional experiments and qualitative analysis further support the potential of SEEA-R1 for future research in scalable embodied intelligence.

cs.AI↗

arXiv2025

WoW: Towards a World omniscient World model Through Embodied Interaction

Humans develop an understanding of intuitive physics through active interaction with the world. This approach is in stark contrast to current video models, such as Sora, which rely on passive observation and therefore struggle with grasping physical causality. This observation leads to our central hypothesis: authentic physical intuition of the world model must be grounded in extensive, causally rich interactions with the real world. To test this hypothesis, we present WoW, a 14-billion-parameter generative world model trained on 2 million robot interaction trajectories. Our findings reveal that the model's understanding of physics is a probabilistic distribution of plausible outcomes, leading to stochastic instabilities and physical hallucinations. Furthermore, we demonstrate that this emergent capability can be actively constrained toward physical realism by SOPHIA, where vision-language model agents evaluate the DiT-generated output and guide its refinement by iteratively evolving the language instructions. In addition, a co-trained Inverse Dynamics Model translates these refined plans into executable robotic actions, thus closing the imagination-to-action loop. We establish WoWBench, a new benchmark focused on physical consistency and causal reasoning in video, where WoW achieves state-of-the-art performance in both human and autonomous evaluation, demonstrating strong ability in physical causality, collision dynamics, and object permanence. Our work provides systematic evidence that large-scale, real-world interaction is a cornerstone for developing physical intuition in AI. Models, data, and benchmarks will be open-sourced.

cs.RO↗

arXiv2025

ReWatch-R1: Boosting Complex Video Reasoning in Large Vision-Language Models through Agentic Data Synthesis

While Reinforcement Learning with Verifiable Reward (RLVR) significantly advances image reasoning in Large Vision-Language Models (LVLMs), its application to complex video reasoning remains underdeveloped. This gap stems primarily from a critical data bottleneck: existing datasets lack the challenging, multi-hop questions and high-quality, video-grounded Chain-of-Thought (CoT) data necessary to effectively bootstrap RLVR. To address this, we introduce ReWatch, a large-scale dataset built to foster advanced video reasoning. We propose a novel multi-stage synthesis pipeline to synthesize its three components: ReWatch-Caption, ReWatch-QA, and ReWatch-CoT. A core innovation is our Multi-Agent ReAct framework for CoT synthesis, which simulates a human-like "re-watching" process to generate video-grounded reasoning traces by explicitly modeling information retrieval and verification. Building on this dataset, we develop ReWatch-R1 by post-training a strong baseline LVLM with Supervised Fine-Tuning (SFT) and our RLVR framework. This framework incorporates a novel Observation \& Reasoning (O\&R) reward mechanism that evaluates both the final answer's correctness and the reasoning's alignment with video content, directly penalizing hallucination. Our experiments show that ReWatch-R1 achieves state-of-the-art average performance on five challenging video reasoning benchmarks. Project Page: https://rewatch-r1.github.io

cs.CV↗

arXiv2025

MARS2 2025 Challenge on Multimodal Reasoning: Datasets, Methods, Results, Discussion, and Outlook

This paper reviews the MARS2 2025 Challenge on Multimodal Reasoning. We aim to bring together different approaches in multimodal machine learning and LLMs via a large benchmark. We hope it better allows researchers to follow the state-of-the-art in this very dynamic area. Meanwhile, a growing number of testbeds have boosted the evolution of general-purpose large language models. Thus, this year's MARS2 focuses on real-world and specialized scenarios to broaden the multimodal reasoning applications of MLLMs. Our organizing team released two tailored datasets Lens and AdsQA as test sets, which support general reasoning in 12 daily scenarios and domain-specific reasoning in advertisement videos, respectively. We evaluated 40+ baselines that include both generalist MLLMs and task-specific models, and opened up three competition tracks, i.e., Visual Grounding in Real-world Scenarios (VG-RS), Visual Question Answering with Spatial Awareness (VQA-SA), and Visual Reasoning in Creative Advertisement Videos (VR-Ads). Finally, 76 teams from the renowned academic and industrial institutions have registered and 40+ valid submissions (out of 1200+) have been included in our ranking lists. Our datasets, code sets (40+ baselines and 15+ participants' methods), and rankings are publicly available on the MARS2 workshop website and our GitHub organization page https://github.com/mars2workshop/, where our updates and announcements of upcoming events will be continuously provided.

cs.CV↗

arXiv2025

C3DE: Causal-Aware Collaborative Neural Controlled Differential Equation for Long-Term Urban Crowd Flow Prediction

Long-term urban crowd flow prediction suffers significantly from cumulative sampling errors, due to increased sequence lengths and sampling intervals, which inspired us to leverage Neural Controlled Differential Equations (NCDEs) to mitigate this issue. However, regarding the crucial influence of Points of Interest (POIs) evolution on long-term crowd flow, the multi-timescale asynchronous dynamics between crowd flow and POI distribution, coupled with latent spurious causality, poses challenges to applying NCDEs for long-term urban crowd flow prediction. To this end, we propose Causal-aware Collaborative neural CDE (C3DE) to model the long-term dynamic of crowd flow. Specifically, we introduce a dual-path NCDE as the backbone to effectively capture the asynchronous evolution of collaborative signals across multiple time scales. Then, we design a dynamic correction mechanism with the counterfactual-based causal effect estimator to quantify the causal impact of POIs on crowd flow and minimize the accumulation of spurious correlations. Finally, we leverage a predictor for long-term prediction with the fused collaborative signals of POI and crowd flow. Extensive experiments on three real-world datasets demonstrate the superior performance of C3DE, particularly in cities with notable flow fluctuations.

cs.LG↗

arXiv2025

Planar Turán number of disjoint union of $C_3$ and $C_5$

The planar Turán number of $H$, denoted by $ex_{\mathcal{P}}(n,H)$, is the maximum number of edges in an $n$-vertex $H$-free planar graph. The planar Turán number of $k\geq 3$ vertex-disjoint union of cycles is the trivial value $3n-6$. Let $C_{\ell}$ denote the cycle of length $\ell$ and $C_{\ell}\cup C_t$ denote the union of disjoint cycles $C_{\ell}$ and $C_t$. The planar Turán number $ex_{\mathcal{P}}(n,H)$ is known if $H=C_{\ell}\cup C_k$, where $\ell,k\in \{3,4\}$. In this paper, we determine the value $ex_{\mathcal{P}}(n,C_3\cup C_5)=\lfloor\frac{8n-13}{3}\rfloor$ and characterize the extremal graphs when $n$ is sufficiently large.

math.CO↗

arXiv2025

Atmospheric Turbulence-Resilient Long-Range Fourier Ptychography

While Fourier ptychography (FP) offers super-resolution for macroscopic imaging, its real-world application is severely hampered by atmospheric turbulence, a challenge largely unaddressed in existing macroscopic FP research operating under idealized conditions. This work establishes, to our knowledge, the first comprehensive computational framework specifically designed for turbulence mitigation in long-range FP, termed Turbulence-Mitigated FP (TMFP). Rather than correcting pupil errors, an image degradation model is developed alongside a reconstruction pipeline inspired by speckle interferometry. By taking multiple short-exposure randomly-distorted measurements and exploiting their statistical properties, the diffraction-limited sub-aperture images can be recovered for further FP reconstruction. Numerical simulations and experimental validations under optical turbulence demonstrate the method's robustness, resolution enhancement, and practicality in adverse conditions, paving the way for the reliable deployment of high-resolution macroscopic FP in real-world scenarios.

physics.optics↗

arXiv2025

Planar Turán number of two adjacent cycles

The planar Turán number of $H$, denoted by $ex_{\mathcal{P}}(n,H)$, is the maximum number of edges in an $n$-vertex $H$-free planar graph. The planar Turán number of $k(k\geq 3)$ vertex-disjoint union of cycles is the trivial value $3n-6$. We determine the planar Turán number of $C_{3}\text{-}C_{3}$ and $C_{3}\text{-}C_{4}$, where $C_{k}\text{-}C_{\ell}$ denotes the graph consisting of two disjoint cycles $C_k$ with an edge connecting them.

math.CO↗

arXiv2025

DLF: Disentangled-Language-Focused Multimodal Sentiment Analysis

Multimodal Sentiment Analysis (MSA) leverages heterogeneous modalities, such as language, vision, and audio, to enhance the understanding of human sentiment. While existing models often focus on extracting shared information across modalities or directly fusing heterogeneous modalities, such approaches can introduce redundancy and conflicts due to equal treatment of all modalities and the mutual transfer of information between modality pairs. To address these issues, we propose a Disentangled-Language-Focused (DLF) multimodal representation learning framework, which incorporates a feature disentanglement module to separate modality-shared and modality-specific information. To further reduce redundancy and enhance language-targeted features, four geometric measures are introduced to refine the disentanglement process. A Language-Focused Attractor (LFA) is further developed to strengthen language representation by leveraging complementary modality-specific information through a language-guided cross-attention mechanism. The framework also employs hierarchical predictions to improve overall accuracy. Extensive experiments on two popular MSA datasets, CMU-MOSI and CMU-MOSEI, demonstrate the significant performance gains achieved by the proposed DLF framework. Comprehensive ablation studies further validate the effectiveness of the feature disentanglement module, language-focused attractor, and hierarchical predictions. Our code is available at https://github.com/pwang322/DLF.

cs.LG↗

arXiv2025

Quantum light sources with configurable lifetime leveraging parity-time symmetry

Quantum light sources with configurable photon lifetimes are essential for large-scale quantum circuits, enabling applications in programmable quantum computing, various quantum key distribution protocols, and quantum tomography techniques. However, the fundamental trade-off between efficiency and photon lifetime imposes significant challenges on the design of high-performance large configurable lifetime quantum light sources. Here, we report on such chip-scale quantum light sources by harnessing the unique feature of parity-time (PT) symmetry. The core design centers on employing PT-symmetric coupling between two microresonators of distinct circumferences, enabling broad-range and selective tuning of intracavity photon density of states. By controlling the alignment between resonators, we achieved a 38-fold photon lifetime tuning range (4 ~ 158 ps), with the shortest lifetimes near the exceptional points of the PT-symmetric systems. The device generates energy-time entangled photon pairs with 87.1 +- 1.1% interference visibility and a heralded second-order autocorrelation of g_h^((2) ) (0)= 0.069 +- 0.001. Our work highlights the potential of PT symmetry for advanced quantum applications, including high-speed communication and programmable quantum computing, quantum coherent tomography, and beyond.

quant-ph↗

arXiv2025

QualityFlow: An Agentic Workflow for Program Synthesis Controlled by LLM Quality Checks

We introduce QualityFlow, a dynamic agentic workflow for program synthesis. Given the English description of a programming problem and a set of unit tests, the model's goal is to synthesize the correct program that solves the problem and passes the tests. QualityFlow includes large language model (LLM) agents resembling a software development team, including code generation, testing, and self-debugging. We propose the LLM Quality Checker, which explicitly "imagines" whether the synthesized programs' execution would conform to the unit tests. The Quality Checks dynamically control the workflow, including actions to submit the final answer, clarify the problem statement, and revert previous workflow steps. Our experiments show that the Quality Checker can precisely accept any correct program, mitigate faulty synthesized tests, and prevent potential workflow deviation. QualityFlow establishes the state-of-the-art results on four program synthesis benchmarks: MBPP, HumanEval, and stricter evaluations from MBPP-EvalPlus and HumanEval-EvalPlus.

cs.SE↗