Searcharxiv⌕ Search

arXiv subjects

Xia Liu

Publications and source records attributed to Xia Liu.

At least 19 recordsLinked to original sources
arXiv2026

SUP-MIMIC: A Multi-Task Clinical Diagnosis Benchmark for Evaluating LLMs' Robustness to Contradictory Evidence

Current evaluations of large language models (LLMs) primarily focus on factual knowledge retrieval, overlooking the fundamental challenge of navigating the complex, non-bijective mappings between clinical indicators and diagnoses. Existing benchmarks fail to assess whether large language models truly possess the reasoning capability required for diagnostic ambiguity scenarios, where identical clinical presentations may correspond to different etiologies, and diagnostic convergence scenarios, where heterogeneous symptoms ultimately indicate the same disease. To address this issue, we propose SUP-MIMIC, a multi-task framework utilizing MIMIC-IV-v3.1 that comprises Basic Assessment (BA), Diagnostic Divergence Task (DDT), and Diagnostic Convergence Task (DCT). Specifically, DDT is designed to evaluate the model's "one-to-many" disambiguation capability among phenotypically similar cases, while DCT assesses the model's ability to identify "many-to-one" diagnostic patterns across different pathophysiological pathways. Comprehensive evaluation of state-of-the-art LLMs reveals substantial performance degradation on DDT and DCT compared to baseline tasks, exposing a systemic reliance on statistical shortcuts over genuine causal reasoning. Our findings further highlight a conservative bias toward "healthy" predictions, implying non-trivial risks for missed diagnoses in realistic medical settings. This work establishes a rigorous methodology for quantifying clinical reasoning robustness and provides a roadmap for enhancing the safety of language models in clinical medicine.

cs.CL↗
arXiv2025

Dynamic LOCC Circuits for Automated Entanglement Manipulation

Due to the limited qubit number of quantum devices, distributed quantum computing is considered a promising pathway to overcome this constraint. In this paradigm, multiple quantum processors are interconnected to form a cohesive computational network, and the most natural set of free operations is local operations and classical communication (LOCC). However, designing a practical LOCC protocol for a particular task has been a tough problem. In this work, we propose a general and flexible framework called dynamic LOCCNet (DLOCCNet) to simulate and design LOCC protocols. We demonstrate its effectiveness in two key applications: entanglement distillation and distributed state discrimination. The protocols designed by DLOCCNet, in contrast to conventional ones, can solve larger-sized problems with reduced training time, making the framework a practical and scalable tool for current quantum devices. This work advances our understanding of the capabilities and limitations of LOCC while providing a powerful methodology for protocol design.

quant-ph↗
arXiv2025

Learning Wisdom from Errors: Promoting LLM's Continual Relation Learning through Exploiting Error Cases

Continual Relation Extraction (CRE) aims to continually learn new emerging relations while avoiding catastrophic forgetting. Existing CRE methods mainly use memory replay and contrastive learning to mitigate catastrophic forgetting. However, these methods do not attach importance to the error cases that can reveal the model's cognitive biases more effectively. To address this issue, we propose an instruction-based continual contrastive tuning approach for Large Language Models (LLMs) in CRE. Different from existing CRE methods that typically handle the training and memory data in a unified manner, this approach splits the training and memory data of each task into two parts respectively based on the correctness of the initial responses and treats them differently through dual-task fine-tuning. In addition, leveraging the advantages of LLM's instruction-following ability, we propose a novel instruction-based contrastive tuning strategy for LLM to continuously correct current cognitive biases with the guidance of previous data in an instruction-tuning manner, which mitigates the gap between old and new relations in a more suitable way for LLMs. We experimentally evaluate our model on TACRED and FewRel, and the results show that our model achieves new state-of-the-art CRE performance with significant improvements, demonstrating the importance of specializing in exploiting error cases.

cs.CL↗
arXiv2025

Trae Agent: An LLM-based Agent for Software Engineering with Test-time Scaling

Software issue resolution is a critical challenge in software engineering and has garnered increasing attention in recent years. With the rapid advancement of large language models (LLMs), substantial progress has been made in addressing real-world software engineering tasks. Recent studies have introduced ensemble reasoning techniques to enhance the performance of LLM-based issue resolution. However, existing prompting-based methods still face limitations in effectively exploring large ensemble spaces and lack the capacity for repository-level understanding, both of which constrain their overall effectiveness. In this paper, we propose Trae Agent, the first agent-based ensemble reasoning approach for repository-level issue resolution. Trae Agent formulates our goal as an optimal solution search problem and addresses two key challenges, i.e., large ensemble spaces and repository-level understanding, through modular agents for generation, pruning, and selection. We conduct extensive experiments using three leading LLMs on the widely-adopted SWE-bench benchmark, comparing Trae Agent against four state-of-the-art ensemble reasoning techniques. Experimental results demonstrate that Trae Agent consistently achieves superior performance, with an average improvement of 10.22% over all baselines in terms of Pass@1. Trae Agent has achieved first place on the SWE-bench Verified leaderboard, with a notable Pass@1 score of 75.20%. We are pleased to release Trae Agent as an open-source project to support the research community, with all resources available at https://github.com/bytedance/trae-agent.

cs.SE↗
arXiv2025

LCQNN: Linear Combination of Quantum Neural Networks

Quantum neural networks combine quantum computing with advanced data-driven methods, offering promising applications in quantum machine learning. However, the optimal paradigm for balancing trainability and expressivity in QNNs remains an open question. To address this, we introduce the Linear Combination of Quantum Neural Networks (LCQNN) framework, which uses the linear combination of unitaries concept to create a tunable design that mitigates vanishing gradients without incurring excessive classical simulability. We show how specific structural choices, such as adopting $k$-local control unitaries or restricting the model to certain group-theoretic subspaces, prevent gradients from collapsing while maintaining sufficient expressivity for complex tasks. We further employ the LCQNN model to handle supervised learning tasks, demonstrating its effectiveness on real datasets. In group action scenarios, we show that by exploiting symmetry and excluding exponentially large irreducible subspaces, the model circumvents barren plateaus. Overall, LCQNN provides a novel framework for focusing quantum resources into architectures that are practically trainable yet expressive enough to tackle challenging machine learning applications.

quant-ph↗
arXiv2025

Use of Air Quality Sensor Network Data for Real-time Pollution-Aware POI Suggestion

This demo paper introduces AirSense-R, a privacy-preserving mobile application that delivers real-time, pollution-aware recommendations for urban points of interest (POIs). By merging live air quality data from AirSENCE sensor networks in Bari (Italy) and Cork (Ireland) with user preferences, the system enables health-conscious decision-making. It employs collaborative filtering for personalization, federated learning for privacy, and a prediction engine to detect anomalies and interpolate sparse sensor data. The proposed solution adapts dynamically to urban air quality while safeguarding user privacy. The code and demonstration video are available at https://github.com/AirtownApp/Airtown-Application.git.

cs.IR↗
arXiv2024

Quantum Entanglement Allocation through a Central Hub

Establishing a fully functional quantum internet relies on the efficient allocation of multipartite entangled states, which enables advanced quantum communication protocols, secure multipartite quantum key distribution, and distributed quantum computing. In this work, we propose local operations and classical communication (LOCC) protocols for allocating generalized $N$-qubit W states within a centralized hub architecture, where the central hub node preshares Bell states with each end node. We develop a detailed analysis of the optimality of the resources required for our proposed W-state allocation protocol and the previously proposed GHZ-state protocol. Our results show that these protocols deterministically and exactly distribute states using only $N$ qubits of quantum memory within the central system, with communication costs of $2N - 2$ and $N$ classical bits for the W and GHZ states, respectively. These resource-efficient LOCC protocols are further proven to be optimal within the centralized hub architecture, outperforming conventional teleportation protocols for entanglement distribution in both memory and communication costs. Our results provide a more resource-efficient method for allocating essential multipartite entangled states in quantum networks, paving the way for the realization of a quantum internet with enhanced efficiency.

quant-ph↗
arXiv2024

RepoMasterEval: Evaluating Code Completion via Real-World Repositories

With the growing reliance on automated code completion tools in software development, the need for comprehensive evaluation benchmarks has become critical. Existing benchmarks focus more on code completion in function and class level by providing text descriptions to prompt the model. By contrast, such descriptive prompt is commonly unavailable in real development and code completion can occur in wider range of situations such as in the middle of a function or a code block. These limitations makes existing evaluation benchmarks poorly align with the practical scenarios of code completion tools. In this paper, we propose RepoMasterEval, a novel benchmark for evaluating code completion models constructed from real-world repositories. Each benchmark datum is generated by masking a code snippet (ground truth) from one source code file with existing test suites. To improve test accuracy of model generated code, we employ mutation testing to measure the effectiveness of the test cases and we manually crafted new test cases for those test suites with low mutation score. Our empirical evaluation on 10 state-of-the-art models shows that test argumentation is critical in improving the accuracy of the benchmark and RepoMasterEval is able to report variance in model performance in real-world scenarios. The deployment of RepoMasterEval also revealed that the benchmark is useful to give accurate feedback during model training and the score is in high correlation with the model's performance in practice.

cs.SE↗
arXiv2024

Optimal unilocal virtual quantum broadcasting

Quantum broadcasting is central to quantum information processing and characterizes the correlations within quantum states. Nonetheless, traditional quantum broadcasting encounters inherent limitations dictated by the principles of quantum mechanics. In a previous study, Parzygnat et al. [Phys. Rev. Lett. 132, 110203 (2024)] introduced a canonical broadcasting quantum map that goes beyond the quantum no-broadcasting theorem through a virtual process. In this work, we generalize the concept of virtual broadcasting to unilocal broadcasting by incorporating a reference system and introduce protocols that can be approximated using physical operations with minimal cost. First, we propose a universal unilocal protocol enabling multiple parties to share the correlations of a target bipartite state, which is encoded in the expectation value for any observable. Second, we formalize the simulation cost of a virtual quantum broadcasting protocol into a semidefinite programming problem. Notably, we propose a specific protocol with optimal simulation cost for the 2-broadcasting scenario, revealing an explicit relationship between simulation cost and the quantum system's dimension. Moreover, we establish upper and lower bounds on the simulation cost of the virtual $n$-broadcasting protocol and demonstrate the convergence of the lower bound to the upper bound as the quantum system's dimension increases.

quant-ph↗
arXiv2023

A New Global Optimization Method Based on Simplex Branching for Solving a Class of Non-Convex QCQP Problems

Quadratic constrained quadratic programming problems often occur in various fields such as engineering practice, management science, and network communication. This article mainly studies a non convex quadratic programming problem with convex quadratic constraints. Firstly, based on our existing results, the problem is reconstructed as an equivalent problem with a simple concave quadratic objective function in the result space, with a convex feasible domain. A global optimization algorithm for solving equivalent problems is proposed based on a branch and bound framework that can ensure the global optimality of the solution. This algorithm combines effective relaxation processes with branching processes related to new external approximation techniques. Finally, the theoretical feasibility of the algorithm was analyzed.

math.OC↗
arXiv2023

Minimizing CNOT-count in quantum circuit of the extended Shor's algorithm for ECDLP

Since the elliptic curve discrete logarithms problem (ECDLP) was proposed, it has been widely used in cryptosystem because of its strong security. Although the proposal of the extended Shor's algorithm offers hope for cracking ECDLP, it is debatable whether the algorithm can actually pose a threat in practice. From the perspective of the quantum circuit of the algorithm, we analyze the feasibility of cracking ECDLP with improved quantum circuits using an ion trap quantum computer. We give precise quantum circuits for extended Shor's algorithm to calculate discrete logarithms on elliptic curves over prime fields, including modulus subtraction, three different modulus multiplication, modulus inverse, and windowed arithmetic. Whereas previous studies mostly focused on minimizing the number of qubits or the depth of the circuit, we minimize the number of CNOTs, which greatly affects the time to run the algorithm on an ion trap quantum computer. First, we give the implementation of the basic arithmetic with the lowest known number of CNOTs and the construction of an improved modular inverse, point addition, and the windowing technique. Then, we precisely estimate the number of improved quantum circuits needed to perform the extended Shor's algorithm for factoring an n-bit integer. We analyze the running time and feasibility of the extended Shor's algorithm on an ion trap quantum computer according to the number of CNOTs. Finally, we discussed the lower bound of the number of CNOTs needed to implement the extended Shor's algorithm.

quant-ph↗
arXiv2023

Linker Code Size Optimization for Native Mobile Applications

Modern mobile applications have grown rapidly in binary size, which restricts user growth and hinders updates for existing users. Thus, reducing the binary size is important for application developers. Recent studies have shown the possibility of using link-time code size optimizations by re-invoking certain compiler optimizations on the linked intermediate representation of the program. However, such methods often incur significant build time overhead and require intrusive changes to the existing build pipeline. In this paper, we propose several novel optimization techniques that do not require significant customization to the build pipeline and reduce binary size with low build time overhead. As opposed to re-invoking the compiler during link time, we perform true linker optimization directly as optimization passes within the linker. This enables more optimization opportunities such as pre-compiled libraries that prior work often could not optimize. We evaluate our techniques on several commercial iOS applications including NewsFeedApp, ShortVideoApp, and CollaborationSuiteApp, each with hundreds of millions of daily active users. Our techniques on average achieve 18.4% binary size reduction across the three commercial applications without any user-perceivable performance degradations.

cs.NI↗
arXiv2022

Strings And Colorings Of Topological Coding Towards Asymmetric Topology Cryptography

We, for anti-quantum computing, will discuss various number-based strings, such as number-based super-strings, parameterized strings, set-based strings, graph-based strings, integer-partitioned and integer-decomposed strings, Hanzi-based strings, as well as algebraic operations based on number-based strings. Moreover, we introduce number-based string-colorings, magic-constraint colorings, and vector-colorings and set-colorings related with strings. For the technique of encrypting the entire network at once, we propose graphic lattices related with number-based strings, Hanzi-graphic lattices, string groups, all-tree-graphic lattices. We study some topics of asymmetric topology cryptography, such as topological signatures, Key-pair graphs, Key-pair strings, one-encryption one-time and self-certification algorithms. Part of topological techniques and algorithms introduced here are closely related with NP-complete problems or NP-hard problems.

cs.IT↗
arXiv2022

Mitigating barren plateaus of variational quantum eigensolvers

Variational quantum algorithms (VQAs) are expected to establish valuable applications on near-term quantum computers. However, recent works have pointed out that the performance of VQAs greatly relies on the expressibility of the ansatzes and is seriously limited by optimization issues such as barren plateaus (i.e., vanishing gradients). This work proposes the state efficient ansatz (SEA) for accurate ground state preparation with improved trainability. We show that the SEA can generate an arbitrary pure state with much fewer parameters than a universal ansatz, making it efficient for tasks like ground state estimation. Then, we prove that barren plateaus can be efficiently mitigated by the SEA and the trainability can be further improved most quadratically by flexibly adjusting the entangling capability of the SEA. Finally, we investigate a plethora of examples in ground state estimation where we obtain significant improvements in the magnitude of cost gradient and the convergence speed.

quant-ph↗
arXiv2021

CNOT-count optimized quantum circuit of the Shor's algorithm

We present improved quantum circuit for modular exponentiation of a constant, which is the most expensive operation in Shor's algorithm for integer factorization. While previous work mostly focuses on minimizing the number of qubits or the depth of circuit, we try to minimize the number of CNOT gate which primarily determines the running time on a ion trap quantum computer. First, we give the implementation of basic arithmetic with known lowest number of CNOT gate and the construction of improved modular exponentiation of a constant by accumulating intermediate date and windowing technique. Then, we precisely estimate the number of improved quantum circuit to perform Shor's algorithm for factoring a $n$-bit integer, which is $217\frac{n^3}{\log_2n}+4n^2+n$. According to the number of CNOT gates, we analyze the running time and feasibility of Shor's algorithm on a ion trap quantum computer. Finally, we discuss the lower bound of CNOT numbers needed to implement Shor's algorithm.

quant-ph↗
arXiv2021

Quantum Identity-Based Encryption from the Learning with Errors Problem

In order to prevent eavesdropping and tampering, the network security protocols use a handshake with an asymmetric cipher to establish a session-specific shared key with which further communication is encrypted using a symmetric cipher. The commonly used asymmetric algorithms include public key encryption, key exchange and identity-based encryption(IBE). However, the network security protocols based on classic identity-based encryption do not have perfect forward security. To solve the problem, we construct the first quantum IBE (QIBE) scheme based on the learning with errors problem, and prove that our scheme is fully secure under the random oracle. Moreover, we construct the quantum circuit of our QIBE scheme and give an estimate of the quantum resource of our circuit including the numbers of Hadamard gate, phase gate, T gate, CNOT gate and the total qubits used in the circuit, and conclude that the quantum resources required by our scheme increase linearly with the number of bits of the encrypted quantum plaintext. Our scheme exhibits the following advantages: (i) The classic key generation center (KGC) system still can be used for our QIBE scheme to generate and distribute the secret identity keys so that the cost can be reduced when the scheme is implemented. The reason why the classic KGC can be used is that the public and private keys are in the form of classic bits. (ii) The network security protocols using a handshake with our QIBE scheme can provide perfect forward security. In our scheme, the ciphertext is transmitted in the form of a quantum state that is unknown to the adversary and therefore cannot be copied and stored. Thus, in the network security protocols based on our QIBE construction, the adversary cannot decrypt the previous quantum ciphertext to threat the previous session keys even if the identity secret key is threatened.

cs.CR↗
arXiv2020

Revisiting the distributions of Jupiter's irregular moons: I. physical characteristics

As the identified number of Jupiter's moons has skyrocketed to 79, some of them have been regrouped. In this work, we continue to identify the potential distributions of the physical characteristics of Jupiter's irregular moons. By using nonparametric Kolmogorov-Smirnov tests, we verified more than 20 commonly used distributions and found that surprisingly, almost all the physical characteristics (i.e., the equatorial radius, equatorial circumference, circumference, volume, mass, surface gravity and escape velocity) of the moons in the Ananke and Carme groups follow log-logistic distributions. Additionally, more than half of the physical characteristics of the moons in the Pasiphae group are theoretically subject to this type of distribution. The discovery of an increasing number of Jupiter's irregular moons combined with strict analytical derivations, it is increasingly clear and possible to anticipate that the physical characteristics of most irregular moons follow log-logistic distributions.

astro-ph.EP↗