arXiv subjects

Zhen Yang

Publications and source records attributed to Zhen Yang.

At least 91 records · Page 5Linked to original sources

arXiv2026

TransLibEval: Demystify Large Language Models' Capability in Third-party Library-targeted Code Translation

In recent years, Large Language Models (LLMs) have been widely studied in the code translation field on the method, class, and even repository levels. However, most of these benchmarks are limited in terms of Third-Party Library (TPL) categories and scales, making TPL-related errors hard to expose and hindering the development of targeted solutions. Considering the high dependence (over 90%) on TPLs in practical programming, demystifying and analyzing LLMs' code translation performance involving various TPLs becomes imperative. To address this gap, we construct TransLibEval, the first benchmark dedicated to library-centric code translation. It consists of 200 real-world tasks across Python, Java, and C++, each explicitly involving TPLs from diverse categories such as data processing, machine learning, and web development, with comprehensive dependency coverage and high-coverage test suites. We evaluate seven recent LLMs of commercial, general, and code-specialized families under six translation strategies of three categories: Direct, IR-guided, and Retrieval-augmented. Experimental results show a dramatic performance drop compared with library-free settings (average CA decline over 60%), while diverse strategies demonstrate heterogeneous advantages. Furthermore, we analyze 4,831 failed cases from GPT-4o, one of the State-of-the-Art (SOTA) LLMs, revealing numerous third-party reference errors that were obscured previously. These findings highlight the unique challenges of library-centric translation and provide practical guidance for improving TPL-aware code intelligence.

cs.SE↗

arXiv2026

Dressed-state relaxation in coupled qubits as a source of two-qubit gate errors

Understanding error mechanisms in two-qubit gate operations is essential for building high-fidelity quantum processors. While prior studies predominantly treat dephasing noise as either Markovian or predominantly low-frequency, realistic qubit environments exhibit structured, frequency-dependent spectra. Here we demonstrate that noise at frequencies matching the dressed-state energy splitting--set by the inter-qubit coupling strength g--induces a distinct relaxation channel that degrades gate performance. Through combined theoretical analysis and experimental verification on superconducting qubits with engineered noise spectra, we show that two-qubit gate errors scale predictably with the noise power spectral density at frequency 2g, extending the concept of $T_{1ρ}$ relaxation to interacting systems. This frequency-selective relaxation mechanism, universal across platforms, enriches our understanding of decoherence pathways during gate operations. The same mechanism sets coherence limits for dual-rail or singlet-triplet encodings.

quant-ph↗

arXiv2026

Suppressing spurious transitions using spectrally balanced pulse

Achieving precise control over quantum systems presents a significant challenge, especially in many-body setups, where residual couplings and unintended transitions undermine the accuracy of quantum operations. In superconducting qubits, parasitic interactions -- both between distant qubits and with spurious two-level systems -- can severely limit the performance of quantum gates. In this work, we introduce a pulse-shaping technique that uses spectrally balanced microwave pulses to suppress undesired transitions. Experimental results demonstrate an order-of-magnitude reduction in spurious excitations between weakly detuned qubits, as well as a substantial decrease in single-qubit gate errors caused by a strongly coupled two-level defect over a broad frequency range. Our method provides a simple yet powerful solution to mitigate adverse effects from parasitic couplings, enhancing the fidelity of quantum operations and expanding feasible frequency allocations for large-scale quantum devices.

quant-ph↗

arXiv2026

Advantages and disadvantages of maximally entangled states in dilaton black hole background

We investigate quantum entanglement and coherence for four classes of Bell-like fermionic states in the vicinity of the event horizon of a Garfinkle-Horowitz-Strominger (GHS) dilaton black hole. Contrary to the common expectation that maximally entangled states always provide superior quantum resources, our results show that their entanglement can be lower than that of suitably chosen non-maximally entangled states in this curved spacetime background. This reveals that non-maximally entangled states may offer operational advantages for entanglement-based tasks under gravitational effects. In contrast, quantum coherence exhibits monotonic behavior: larger initial coherence leads to systematically enhanced robustness against the dilaton induced degradation. These results indicate that the optimal choice of initial quantum states depends sensitively on the specific quantum resource, either quantum entanglement or quantum coherence, required for quantum information processing near a dilaton black hole.

gr-qc↗

arXiv2026

Mitigating Measurement Crosstalk via Pulse Shaping

Quantum error correction protocols require rapid and repeated qubit measurements. While multiplexed readout in superconducting quantum systems improves efficiency, fast probe pulses introduce spectral broadening, leading to signal leakage into neighboring readout resonators. This crosstalk results in qubit dephasing and degraded readout fidelity. Here, we introduce a pulse shaping technique inspired by the derivative removal by adiabatic gate (DRAG) protocol to suppress measurement crosstalk during fast readout. By engineering a spectral notch at neighboring resonator frequencies, the method effectively mitigates spurious signal interference. Our approach integrates seamlessly with existing readout architectures, enabling fast, low-crosstalk multiplexed measurements without additional hardware overhead - a critical advancement for scalable quantum computing.

quant-ph↗

arXiv2026

TransMamba: A Sequence-Level Hybrid Transformer-Mamba Language Model

Transformers are the cornerstone of modern large language models, but their quadratic computational complexity limits efficiency in long-sequence processing. Recent advancements in Mamba, a state space model (SSM) with linear complexity, offer promising efficiency gains but suffer from unstable contextual learning and multitask generalization. Some works conduct layer-level hybrid structures that combine Transformer and Mamba layers, aiming to make full use of both advantages. This paper proposes TransMamba, a novel sequence-level hybrid framework that unifies Transformer and Mamba through shared parameter matrices (QKV and CBx), and thus could dynamically switch between attention and SSM mechanisms at different token lengths and layers. We design the Memory Converter to bridge Transformer and Mamba by converting attention outputs into SSM-compatible states, ensuring seamless information flow at TransPoints where the transformation happens. The TransPoint scheduling is also thoroughly explored for balancing effectiveness and efficiency. We conducted extensive experiments demonstrating that TransMamba achieves superior training efficiency and performance compared to single and hybrid baselines, and validated the deeper consistency between Transformer and Mamba paradigms at sequence level, offering a scalable solution for next-generation language modeling. Code and data are available at https://github.com/Yixing-Li/TransMamba

cs.LG↗

arXiv2026

Can LLMs See Without Pixels? Benchmarking Spatial Intelligence from Textual Descriptions

Recent advancements in Spatial Intelligence (SI) have predominantly relied on Vision-Language Models (VLMs), yet a critical question remains: does spatial understanding originate from visual encoders or the fundamental reasoning backbone? Inspired by this question, we introduce SiT-Bench, a novel benchmark designed to evaluate the SI performance of Large Language Models (LLMs) without pixel-level input, comprises over 3,800 expert-annotated items across five primary categories and 17 subtasks, ranging from egocentric navigation and perspective transformation to fine-grained robotic manipulation. By converting single/multi-view scenes into high-fidelity, coordinate-aware textual descriptions, we challenge LLMs to perform symbolic textual reasoning rather than visual pattern matching. Evaluation results of state-of-the-art (SOTA) LLMs reveals that while models achieve proficiency in localized semantic tasks, a significant "spatial gap" remains in global consistency. Notably, we find that explicit spatial reasoning significantly boosts performance, suggesting that LLMs possess latent world-modeling potential. Our proposed dataset SiT-Bench serves as a foundational resource to foster the development of spatially-grounded LLM backbones for future VLMs and embodied agents. Our code and benchmark will be released at https://github.com/binisalegend/SiT-Bench .

cs.CV↗

arXiv2026

D$^3$R-DETR: DETR with Dual-Domain Density Refinement for Tiny Object Detection in Aerial Images

Detecting tiny objects plays a vital role in remote sensing intelligent interpretation, as these objects often carry critical information for downstream applications. However, due to the extremely limited pixel information and significant variations in object density, mainstream Transformer-based detectors often suffer from slow convergence and inaccurate query-object matching. To address these challenges, we propose D$^3$R-DETR, a novel DETR-based detector with Dual-Domain Density Refinement. By fusing spatial and frequency domain information, our method refines low-level feature maps and utilizes their rich details to predict more accurate object density map, thereby guiding the model to precisely localize tiny objects. Extensive experiments on the AI-TOD-v2 dataset demonstrate that D$^3$R-DETR outperforms existing state-of-the-art detectors for tiny object detection.

cs.CV↗

arXiv2026

SpatialBench: Can Agents Analyze Real-World Spatial Biology Data?

Spatial transcriptomics assays are rapidly increasing in scale and complexity, making computational analysis a major bottleneck in biological discovery. Although frontier AI agents have improved dramatically at software engineering and general data analysis, it remains unclear whether they can extract biological insight from messy, real-world spatial datasets. We introduce SpatialBench, a benchmark of 146 verifiable problems derived from practical spatial analysis workflows spanning five spatial technologies and seven task categories. Each problem provides a snapshot of experimental data immediately prior to an analysis step and a deterministic grader that evaluates recovery of a key biological result. Benchmark data on frontier models shows that base model accuracy remains low (20-38% across model families), with strong model-task and model-platform interactions. Harness design has a large empirical effect on performance, indicating that tools, prompts, control flow, and execution environment should be evaluated and improved as first-class objects. SpatialBench serves both as a measurement tool and a diagnostic lens for developing agents that can interact with real spatial datasets faithfully, transparently, and reproducibly.

cs.AI↗

arXiv2026

GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning

We present GLM-4.1V-Thinking, GLM-4.5V, and GLM-4.6V, a family of vision-language models (VLMs) designed to advance general-purpose multimodal understanding and reasoning. In this report, we share our key findings in the development of the reasoning-centric training framework. We first develop a capable vision foundation model with significant potential through large-scale pre-training, which arguably sets the upper bound for the final performance. We then propose Reinforcement Learning with Curriculum Sampling (RLCS) to unlock the full potential of the model, leading to comprehensive capability enhancement across a diverse range of tasks, including STEM problem solving, video understanding, content recognition, coding, grounding, GUI-based agents, and long document interpretation. In a comprehensive evaluation across 42 public benchmarks, GLM-4.5V achieves state-of-the-art performance on nearly all tasks among open-source models of similar size, and demonstrates competitive or even superior results compared to closed-source models such as Gemini-2.5-Flash on challenging tasks including Coding and GUI Agents. Meanwhile, the smaller GLM-4.1V-9B-Thinking remains highly competitive-achieving superior results to the much larger Qwen2.5-VL-72B on 29 benchmarks. We open-source both GLM-4.1V-9B-Thinking and GLM-4.5V. We further introduce the GLM-4.6V series, open-source multimodal models with native tool use and a 128K context window. A brief overview is available at https://z.ai/blog/glm-4.6v. Code, models and more information are released at https://github.com/zai-org/GLM-V.

cs.CV↗

arXiv2025

StereoPilot: Learning Unified and Efficient Stereo Conversion via Generative Priors

The rapid growth of stereoscopic displays, including VR headsets and 3D cinemas, has led to increasing demand for high-quality stereo video content. However, producing 3D videos remains costly and complex, while automatic Monocular-to-Stereo conversion is hindered by the limitations of the multi-stage ``Depth-Warp-Inpaint'' (DWI) pipeline. This paradigm suffers from error propagation, depth ambiguity, and format inconsistency between parallel and converged stereo configurations. To address these challenges, we introduce UniStereo, the first large-scale unified dataset for stereo video conversion, covering both stereo formats to enable fair benchmarking and robust model training. Building upon this dataset, we propose StereoPilot, an efficient feed-forward model that directly synthesizes the target view without relying on explicit depth maps or iterative diffusion sampling. Equipped with a learnable domain switcher and a cycle consistency loss, StereoPilot adapts seamlessly to different stereo formats and achieves improved consistency. Extensive experiments demonstrate that StereoPilot significantly outperforms state-of-the-art methods in both visual fidelity and computational efficiency. Project page: https://hit-perfect.github.io/StereoPilot/.

cs.CV↗

arXiv2025

Parameter-Efficient Fine-Tuning with Attributed Patch Semantic Graph for Automated Patch Correctness Assessment

Automated program repair (APR) aims to automatically repair program errors without human intervention, and recent years have witnessed a growing interest on this research topic. While much progress has been made and techniques originating from different disciplines have been proposed, APR techniques generally suffer from the patch overfitting issue, i.e., the generated patches are not genuinely correct despite they pass the employed tests. To alleviate this issue, many research efforts have been devoted for automated patch correctness assessment (APCA). In particular, with the emergence of large language model (LLM) technology, researchers have employed LLM to assess the patch correctness and have obtained the state-of-the-art performance. The literature on APCA has demonstrated the importance of capturing patch semantic and explicitly considering certain code attributes in predicting patch correctness. However, existing LLM-based methods typically treat code as token sequences and ignore the inherent formal structure for code, making it difficult to capture the deep patch semantics. Moreover, these LLM-based methods also do not explicitly account for enough code attributes. To overcome these drawbacks, we in this paper design a novel patch graph representation named attributed patch semantic graph (APSG), which adequately captures the patch semantic and explicitly reflects important patch attributes. To effectively use graph information in APSG, we accordingly propose a new parameter-efficient fine-tuning (PEFT) method of LLMs named Graph-LoRA. Extensive evaluations have been conducted to evaluate our method, and the results show that compared to the state-of-the-art methods, our method improves the accuracy and F1 score by 3.1\% to 7.5\% and 3.0\% to 7.1\% respectively.

cs.SE↗

arXiv2025

The FACTS Leaderboard: A Comprehensive Benchmark for Large Language Model Factuality

We introduce The FACTS Leaderboard, an online leaderboard suite and associated set of benchmarks that comprehensively evaluates the ability of language models to generate factually accurate text across diverse scenarios. The suite provides a holistic measure of factuality by aggregating the performance of models on four distinct sub-leaderboards: (1) FACTS Multimodal, which measures the factuality of responses to image-based questions; (2) FACTS Parametric, which assesses models' world knowledge by answering closed-book factoid questions from internal parameters; (3) FACTS Search, which evaluates factuality in information-seeking scenarios, where the model must use a search API; and (4) FACTS Grounding (v2), which evaluates whether long-form responses are grounded in provided documents, featuring significantly improved judge models. Each sub-leaderboard employs automated judge models to score model responses, and the final suite score is an average of the four components, designed to provide a robust and balanced assessment of a model's overall factuality. The FACTS Leaderboard Suite will be actively maintained, containing both public and private splits to allow for external participation while guarding its integrity. It can be found at https://www.kaggle.com/benchmarks/google/facts .

cs.CL↗

arXiv2025

MathSight: A Benchmark Exploring Have Vision-Language Models Really Seen in University-Level Mathematical Reasoning?

Recent advances in Vision-Language Models (VLMs) have achieved impressive progress in multimodal mathematical reasoning. Yet, how much visual information truly contributes to reasoning remains unclear. Existing benchmarks report strong overall performance but seldom isolate the role of the image modality, leaving open whether VLMs genuinely leverage visual understanding or merely depend on linguistic priors. To address this, we present MathSight, a university-level multimodal mathematical reasoning benchmark designed to disentangle and quantify the effect of visual input. Each problem includes multiple visual variants -- original, hand-drawn, photo-captured -- and a text-only condition for controlled comparison. Experiments on state-of-the-art VLMs reveal a consistent trend: the contribution of visual information diminishes with increasing problem difficulty. Remarkably, Qwen3-VL without any image input surpasses both its multimodal variants and GPT-5, underscoring the need for benchmarks like MathSight to advance genuine vision-grounded reasoning in future models.

cs.CV↗

arXiv2025

RoboTidy : A 3D Gaussian Splatting Household Tidying Benchmark for Embodied Navigation and Action

Household tidying is an important application area, yet current benchmarks neither model user preferences nor support mobility, and they generalize poorly, making it hard to comprehensively assess integrated language-to-action capabilities. To address this, we propose RoboTidy, a unified benchmark for language-guided household tidying that supports Vision-Language-Action (VLA) and Vision-Language-Navigation (VLN) training and evaluation. RoboTidy provides 500 photorealistic 3D Gaussian Splatting (3DGS) household scenes (covering 500 objects and containers) with collisions, formulates tidying as an "Action (Object, Container)" list, and supplies 6.4k high-quality manipulation demonstration trajectories and 1.5k naviagtion trajectories to support both few-shot and large-scale training. We also deploy RoboTidy in the real world for object tidying, establishing an end-to-end benchmark for household tidying. RoboTidy offers a scalable platform and bridges a key gap in embodied AI by enabling holistic and realistic evaluation of language-guided robots.

cs.RO↗

arXiv2025

Beyond Flatlands: Unlocking Spatial Intelligence by Decoupling 3D Reasoning from Numerical Regression

Existing Vision Language Models (VLMs) architecturally rooted in "flatland" perception, fundamentally struggle to comprehend real-world 3D spatial intelligence. This failure stems from a dual-bottleneck: input-stage conflict between computationally exorbitant geometric-aware encoders and superficial 2D-only features, and output-stage misalignment where discrete tokenizers are structurally incapable of producing precise, continuous numerical values. To break this impasse, we introduce GEODE (Geometric-Output and Decoupled-Input Engine), a novel architecture that resolves this dual-bottleneck by decoupling 3D reasoning from numerical generation. GEODE augments main VLM with two specialized, plug-and-play modules: Decoupled Rationale Module (DRM) that acts as spatial co-processor, aligning explicit 3D data with 2D visual features via cross-attention and distilling spatial Chain-of-Thought (CoT) logic into injectable Rationale Tokens; and Direct Regression Head (DRH), an "Embedding-as-Value" paradigm which routes specialized control tokens to a lightweight MLP for precise, continuous regression of scalars and 3D bounding boxes. The synergy of these modules allows our 1.5B parameter model to function as a high-level semantic dispatcher, achieving state-of-the-art spatial reasoning performance that rivals 7B+ models.

cs.CV↗

arXiv2025

A 1.431-Competitive Algorithm for Combinatorial Group Testing

In the context of fault-detection problems, the objective is to identify all defective items among a set of $n$ binary-state items using the minimum number of tests. The {group testing} paradigm, which allows testing a subset of items in a single test, serves as a fundamental technique for efficiently classifying large populations. We study a central problem in the combinatorial group testing model where the number $d$ of defective items is unknown in advance. Let $M_α(d|n)$ denote the maximum number of tests required by an algorithm $α$ for this problem, and $M(d,n)$ denote the minimum number of tests required in the worst case when $d$ is known in advance. An algorithm $α$ is called a $c$-\emph{competitive algorithm} if there exist constants $c$ and $a$ such that, for $0\le d < n$, $M_α(d|n)\le cM(d,n)+a$. We design a new adaptive algorithm with a competitive constant $c \le 1.431$, thus pushing the competitive ratio below the best-known one of $1.452$. To achieve this, we propose a novel solution framework based on an unexplored up-zig-zag strategy and a studied strongly competitive algorithm.

math.CO↗

arXiv2025

PlotGen-Bench: Evaluating VLMs on Generating Visualization Code from Diverse Plots across Multiple Libraries

Recent advances in vision-language models (VLMs) have expanded their multimodal code generation capabilities, yet their ability to generate executable visualization code from plots, especially for complex 3D, animated, plot-to-plot transformations, or multi-library scenarios, remains underexplored. To address this gap, we introduce PlotGen-Bench, a comprehensive benchmark for evaluating plot-to-code generation under realistic and complex visualization scenarios. The benchmark spans 9 major categories, 30 subcategories, and 3 core tasks-plot replication, plot transformation, and multi-library generation, covering both 2D, 3D and animated plots across 5 widely used visualization libraries. Through systematic evaluation of state-of-the-art open- and closed-source VLMs, we find that open-source models still lag considerably behind in visual fidelity and semantic consistency, despite achieving comparable code executability. Moreover, all models exhibit substantial degradation on reasoning-intensive tasks such as chart type conversion and animation generation. PlotGen-Bench establishes a rigorous foundation for advancing research toward more capable and reliable VLMs for visualization authoring and code synthesis, with all data and code available at https://plotgen.github.io.

cs.HC↗