arXiv subjects

Xinyu Qiu

Publications and source records attributed to Xinyu Qiu.

17 recordsLinked to original sources

arXiv2026

UI-Venus-2 Technical Report

Multimodal GUI agents have emerged as a promising paradigm for digital task automation, yet transitioning from benchmark-oriented models to dependable real-world applications remains challenging due to limited environment coverage, brittle task construction, and unreliable reward verification. In this work, we present UI-Venus-2, a general-purpose foundation GUI agent designed to operate across mobile, web, and desktop environments through a unified closed-loop reasoning-action framework. To bridge the gap toward practical deployment, we jointly scale three critical dimensions: (1) Environments, expanding coverage to more than 170 multilingual mobile apps and native desktop operating systems; (2) Tasks, employing a deep-research pipeline for function-grounded instruction generation; and (3) Verification, adopting trace-level and sample-level evaluators with visual keypoints and multi-model voting to ensure reliable RL signals for training. Furthermore, we integrate safety-aware mechanisms to ensure controlled execution of consequential actions. By offering a capable, efficient, and open-source foundation, UI-Venus-2 advances the field toward more generalizable, verifiable, and self-reflective agents for real-world applications.

cs.AI↗

arXiv2026

Rank and Range Criteria for Mixed-State Determination from Local Marginals

Determining whether a mixed quantum state is uniquely determined among all states by its k-body marginals (k-UDA) is a fundamental problem in quantum system certification. We develop a range-based approach to this problem by analyzing the structure of the range of the global state. For three-qubit states, we show that states with GHZ-SLOCC-free ranges are 2-UDA at ranks one, three, and four. We derive a necessary and sufficient range criterion for rank-two 2-UDA states and reduce it to a finite quadratic-form test. To cover the remaining range configurations, we formulate an exact range-restricted semidefinite programming criterion and extend it to arbitrary finite-dimensional tripartite states. We also show that every three-qubit state of rank at least five is not 2-UDA, and further extend high-rank obstructions to multipartite systems. For a channel-based multipartite family, we characterize exactly when a state is $(n-1)$-UDA and show that lower-order marginals never suffice. Finally, we apply these results to the certification of genuine multipartite entanglement.

quant-ph↗

arXiv2026

VISTA: View-Consistent Self-Verified Training for GUI Grounding

When applying Group Relative Policy Optimization (GRPO) for GUI Grounding, rollouts are sampled from a single screenshot view; groups often become either all failures on difficult instances or all successes on easy ones, yielding no useful relative advantage. We propose VISTA (View-Consistent Self-Verified Training), a GRPO-based training framework that constructs each comparison group from multiple target-preserving views of the same GUI instance.Each view is generated by a crop that keeps the target element visible and remaps its box exactly, so model rollouts are compared across semantically equivalent but geometrically different inputs. To stabilize short coordinate generation without turning reinforcement learning into unconditional imitation, VISTA further adds a self-verified cross-view anchor: an oracle answer optimized with an advantage-weighted loss, excluded from the group baseline and activated only when the model has produced a maximum-reward rollout. Across five GUI-grounding benchmarks and multiple Qwen backbones, VISTA consistently improves grounding accuracy.On ScreenSpot-Pro, it raises Qwen3-VL 4B/8B/30B-A3B from 55.5/52.7/53.7 to 63.4/65.8/67.0. Robustness analyses further show higher worst-view accuracy and lower prediction flip rates.

cs.AI↗

arXiv2026

UI-Venus-1.5 Technical Report

GUI agents have emerged as a powerful paradigm for automating interactions in digital environments, yet achieving both broad generality and consistently strong task performance remains challenging. In this report, we present UI-Venus-1.5, a unified, end-to-end GUI Agent designed for robust real-world applications. The proposed model family comprises two dense variants (2B and 8B) and one mixture-of-experts variant (30B-A3B) to meet various downstream application scenarios. Compared to our previous version, UI-Venus-1.5 introduces three key technical advances: (1) a comprehensive Mid-Training stage leveraging 10 billion tokens across 30+ datasets to establish foundational GUI semantics; (2) Online Reinforcement Learning with full-trajectory rollouts, aligning training objectives with long-horizon, dynamic navigation in large-scale environments; and (3) a single unified GUI Agent constructed via Model Merging, which synthesizes domain-specific models (grounding, web, and mobile) into one cohesive checkpoint. Extensive evaluations demonstrate that UI-Venus-1.5 establishes new state-of-the-art performance on benchmarks such as ScreenSpot-Pro (69.6%), VenusBench-GD (75.0%), and AndroidWorld (77.6%), significantly outperforming previous strong baselines. In addition, UI-Venus-1.5 demonstrates robust navigation capabilities across a variety of Chinese mobile apps, effectively executing user instructions in real-world scenarios. Code: https://github.com/inclusionAI/UI-Venus; Model: https://huggingface.co/collections/inclusionAI/ui-venus

cs.CV↗

arXiv2026

Unified Generation and Self-Verification for Vision-Language Models via Advantage Decoupled Preference Optimization

Parallel test-time scaling typically trains separate generation and verification models, incurring high training and inference costs. We propose Advantage Decoupled Preference Optimization (ADPO), a unified reinforcement learning framework that jointly learns answer generation and self-verification within a single policy. ADPO introduces two innovations: a preference verification reward improving verification capability and a decoupled optimization mechanism enabling synergistic optimization of generation and verification. Specifically, the preference verification reward computes mean verification scores from positive and negative samples as decision thresholds, providing positive feedback when prediction correctness aligns with answer correctness. Meanwhile, the advantage decoupled optimization computes separate advantages for generation and verification, applies token masks to isolate gradients, and combines masked GRPO objectives, preserving generation quality while calibrating verification scores. ADPO achieves up to +34.1% higher verification AUC and -53.5% lower inference time, with significant gains of +2.8%/+1.4% accuracy on MathVista/MMMU, +1.9 cIoU on ReasonSeg, and +1.7%/+1.0% step success rate on AndroidControl/GUI Odyssey.

cs.CV↗

arXiv2025

Mixed States Uniquely Determined by Marginals and Additivity

Identifying whether a mixed quantum state is uniquely determined among all states (UDA) by its local marginals is a basic problem in quantum information theory. We establish necessary and sufficient conditions under which several classes of multipartite mixed states are UDA by their $k$-partite marginals. We also prove structural properties based on ranges and marginals, and formulate a recursive procedure for the determination of UDA states. We show that sufficiently high rank rules out unique determination from fixed-order marginals, implying that almost all multipartite mixed states are not UDA from such marginals. Finally, we completely characterize the additivity of bipartite UDA states, three-qubit and several families of $n$-qubit product UDA states. These results clarify the boundary between UDA and non-UDA mixed states and provide a framework for local-marginal reconstruction and related certification tasks.

quant-ph↗

arXiv2025

Machine Learning Framework for Efficient Prediction of Quantum Wasserstein Distance

The quantum Wasserstein distance (W-distance) is a fundamental metric for quantifying the distinguishability of quantum operations, with critical applications in quantum error correction. However, computing the W-distance remains computationally challenging for multiqubit systems due to exponential scaling. We present a machine learning framework that efficiently predicts the quantum W-distance by extracting physically meaningful features from quantum state pairs, including Pauli measurements, statistical moments, quantum fidelity, and entanglement measures. Our approach employs both classical neural networks and traditional machine learning models. On three-qubit systems, the best-performing Random Forest model achieves near-perfect accuracy ($R^2 = 0.9999$) with mean absolute errors on the order of $10^{-5}$. We further validate the framework's practical utility by successfully verifying two fundamental theoretical propositions in quantum information theory: the bound on measurement probability differences between unitary operations and the $W_1$ gate error rate bound. The results establish machine learning as a viable and scalable alternative to traditional numerical methods for W-distance computation, with particular promise for real-time quantum circuit assessment and error correction protocol design in NISQ devices.

quant-ph↗

arXiv2025

Performance Review on LLM for solving leetcode problems

This paper presents a comprehensive performance evaluation of Large Language Models (LLMs) in solving programming challenges from Leetcode, a widely used platform for algorithm practice and technical interviews. We began by crawling the Leetcode website to collect a diverse set of problems encompassing various difficulty levels and topics. Using this dataset, we generated solutions with multiple LLMs, including GPT-4 and GPT-3.5-turbo (ChatGPT-turbo). The generated solutions were systematically evaluated for correctness and efficiency. We employed the pass@k metric to assess the success rates within a given number of attempts and analyzed the runtime performance of the solutions. Our results highlight the strengths and limitations of current LLMs [10] in code generation and problem-solving tasks, providing insights into their potential applications and areas for improvement in automated programming assistance.

cs.SE↗

arXiv2025

Zero-Shot End-to-End Relation Extraction in Chinese: A Comparative Study of Gemini, LLaMA and ChatGPT

This study investigates the performance of various large language models (LLMs) on zero-shot end-to-end relation extraction (RE) in Chinese, a task that integrates entity recognition and relation extraction without requiring annotated data. While LLMs show promise for RE, most prior work focuses on English or assumes pre-annotated entities, leaving their effectiveness in Chinese RE largely unexplored. To bridge this gap, we evaluate ChatGPT, Gemini, and LLaMA based on accuracy, efficiency, and adaptability. ChatGPT demonstrates the highest overall performance, balancing precision and recall, while Gemini achieves the fastest inference speed, making it suitable for real-time applications. LLaMA underperforms in both accuracy and latency, highlighting the need for further adaptation. Our findings provide insights into the strengths and limitations of LLMs for zero-shot Chinese RE, shedding light on trade-offs between accuracy and efficiency. This study serves as a foundation for future research aimed at improving LLM adaptability to complex linguistic tasks in Chinese NLP.

cs.CL↗

arXiv2024

Few-Shot Learning with Adaptive Weight Masking in Conditional GANs

Deep learning has revolutionized various fields, yet its efficacy is hindered by overfitting and the requirement of extensive annotated data, particularly in few-shot learning scenarios where limited samples are available. This paper introduces a novel approach to few-shot learning by employing a Residual Weight Masking Conditional Generative Adversarial Network (RWM-CGAN) for data augmentation. The proposed model integrates residual units within the generator to enhance network depth and sample quality, coupled with a weight mask regularization technique in the discriminator to improve feature learning from small-sample categories. This method addresses the core issues of robustness and generalization in few-shot learning by providing a controlled and clear augmentation of the sample space. Extensive experiments demonstrate that RWM-CGAN not only expands the sample space effectively but also enriches the diversity and quality of generated samples, leading to significant improvements in detection and classification accuracy on public datasets. The paper contributes to the advancement of few-shot learning by offering a practical solution to the challenges posed by data scarcity and the need for rapid generalization to new tasks or categories.

cs.CV↗

arXiv2024

Multipartite entangling power by von Neumann entropy

Quantifying the entanglement generation of a multipartite unitary operation is a key problem in quantum information processing. We introduce the definition of multipartite entangling, assisted entangling, and disentangling power, which is a natural generalization of the bipartite ones. We show that they are assumed at a specified quantum state. We analytically derive the entangling power of Schmidt-rank-two multi-qubit unitary operations by the minimal convex sum of modulo-one complex numbers. Besides we show the necessary and sufficient condition that the assisted entangling power of Schmidt-rank-two unitary operations reaches the maximum. We further investigate the widely-used multi-qubit gates, for example, the entangling and assisted entangling power of the $n$-qubit Toffoli gate is one ebit. The entangling power of the three-qubit Fredkin gate is two ebits, and that of the four-qubit Fredkin gate is in two to $\log_25$ ebits.

quant-ph↗

arXiv2024

Deep Reinforcement Learning-based Obstacle Avoidance for Robot Movement in Warehouse Environments

At present, in most warehouse environments, the accumulation of goods is complex, and the management personnel in the control of goods at the same time with the warehouse mobile robot trajectory interaction, the traditional mobile robot can not be very good on the goods and pedestrians to feed back the correct obstacle avoidance strategy, in order to control the mobile robot in the warehouse environment efficiently and friendly to complete the obstacle avoidance task, this paper proposes a deep reinforcement learning based on the warehouse environment, the mobile robot obstacle avoidance Algorithm. Firstly, for the insufficient learning ability of the value function network in the deep reinforcement learning algorithm, the value function network is improved based on the pedestrian interaction, the interaction information between pedestrians is extracted through the pedestrian angle grid, and the temporal features of individual pedestrians are extracted through the attention mechanism, so that we can learn to obtain the relative importance of the current state and the historical trajectory state as well as the joint impact on the robot's obstacle avoidance strategy, which provides an opportunity for the learning of multi-layer perceptual machines afterwards. Secondly, the reward function of reinforcement learning is designed based on the spatial behaviour of pedestrians, and the robot is punished for the state where the angle changes too much, so as to achieve the requirement of comfortable obstacle avoidance; Finally, the feasibility and effectiveness of the deep reinforcement learning-based mobile robot obstacle avoidance algorithm in the warehouse environment in the complex environment of the warehouse are verified through simulation experiments.

cs.RO↗

arXiv2024

Device-independent quantum state discrimination

Quantum state discrimination depicts the general progress of extracting classical information from quantum systems. We show that quantum state discrimination can be realized in a device-independent scenario using tools of self-testing results. That is, the states can be discriminated credibly with the untrusted experiment devices by the correspondence between quantum correlations and states. In detail, we show that two arbitrary states can be discriminated in a device-independent manner when they are not conjugate with each other, while other states can be discriminated measurement-device-independently. To fulfill the device-independent requirement, the measurements are restricted on Pauli observables. The influence of this restriction is acceptable based on the guessing probability analysis for minimum error discrimination.

quant-ph↗

arXiv2023

Quantum Wasserstein distance between unitary operations

Quantifying the effect of noise on unitary operations is an essential task in quantum information processing. We propose the quantum Wasserstein distance between unitary operations, which shows an explanation for quantum circuit complexity and characterizes local distinguishability of multi-qudit operations. We show analytical calculation of the distance between identity and widely-used quantum gates including SWAP, CNOT, and other controlled gates. As an application, we estimate the closeness between quantum gates in circuit, and show that the noisy operation simulates the ideal one well when they become close under the distance. Further we introduce the $W_1$ error rate by the distance, and establish the relation between the $W_1$ error rate and two practical cost measures of recovery operation in quantum error-correction under typical noise scenarios.

quant-ph↗

arXiv2022

Controlled remote implementation of operations via graph states

We propose protocols for controlled remote implementation of operations with convincing control power. Sharing a $(2N+1)$-partite graph state, $2N$ participants collaborate to prepare the stator and realize the operation $\otimes_{j=1}^N\exp{[i\alpha_j\sigma_{n_{O_j}}]}$ on $N$ unknown states for distributed systems $O_j$, with the permission of a controller. All the implementation requirements of our protocol can be satisfied by means of local operations and classical communications, and the experimental feasibility is presented according to current techniques. We characterize the entanglement requirement of our protocol in terms of geometric measure of entanglement. It turns out to be economic to realize the control function from the perspective of entanglement cost. Further we show that the control power of our protocol is reliable by positive operator valued measurement.

quant-ph↗

arXiv2022

Quantum cost of dense coding and teleportation

The quantum cost is a key ingredient to evaluate the quality of quantum protocols from a practical viewpoint. We show that the quantum cost of d-dimensional dense coding protocol is equal to d+3 when transmitting the classical message (0,0), and that is equal to d+4 when transmitting other classical message. It appears linear growth with the dimension and thus makes sense for implementation. In contrast, the quantum cost of high-dimensional teleportation protocols is equal to 13, which is the maximum value of the cost for the two-dimensional case. As an application, we establish the relation between the quantum cost and fidelity of dense coding protocols in terms of four typical noise scenario.

quant-ph↗

arXiv2021

Some special complex Hadamard matrices of order six

The complete classification of $6\times 6$ complex Hadamard matrices (CHMs) is a long-standing open problem. In this paper we investigate a series of CHMs, such as the CHMs containing a $2\times 3$ submatrix with rank one, the CHMs containing exactly three distinct elements and all elements of the first row being one, the $H_2$-reducible matrices containing exactly three distinct matrix elements, and the CHMs containing a $3\times 3$ Hadamard submatrix. We characterize all forms of these CHMs. Our results are the latest progress on the complete classification of CHMs.

math-ph↗