arXiv subjects

Ke Liu

Publications and source records attributed to Ke Liu.

At least 55 records · Page 3Linked to original sources

arXiv2025

SOFT: a high-performance simulator for universal fault-tolerant quantum circuits

Circuit simulation tools are critical for developing and assessing quantum-error-correcting and fault-tolerant strategies. In this work, we present SOFT, a high-performance SimulatOr for universal Fault-Tolerant quantum circuits. Integrating the generalized stabilizer formalism and highly optimized GPU parallelization, SOFT enables the simulation of noisy quantum circuits containing non-Clifford gates at a scale not accessible with existing tools. To provide a concrete demonstration, we simulate the state-of-the-art magic state cultivation (MSC) protocol at code distance $d=5$, involving 42 qubits, 72 $T$ / $T^\dagger$ gates, and mid-circuit measurements. Using only modest GPU resources, SOFT performs over 200 billion shots and achieves the first ground-truth simulation of the cultivation protocol at a non-trivial scale. This endeavor not only certifies the MSC's effectiveness for generating high-fidelity logical $T$-states, but also reveals a large discrepancy between the actual logical error rate and the previously reported values. Our work demonstrates the importance of reliable simulation tools for fault-tolerant architecture design, advancing the field from simulating quantum memory to simulating a universal quantum computer.

quant-ph↗

arXiv2025

Generation of Polarization-Tunable Hybrid Cylindrical Vector gamma Rays

Cylindrical vector (CV) gamma rays can introduce spatially structured polarization as a new degree of freedom for fundamental research and practical applications. However, their generation and control remain largely unexplored. Here, we put forward a novel method to generate CV gamma rays with tunable hybrid polarization via a rotating electron beam interacting with a solid foil. In this process, the beam generates a coherent transition radiation field and subsequently emits gamma rays through nonlinear Compton scattering. By manipulating the initial azimuthal momentum of the beam, the polarization angle of gamma rays relative to the transverse momentum can be controlled, yielding tunable hybrid CV polarization states. Three-dimensional spin-resolved particle-in-cell simulations demonstrate continuous tuning of the polarization angle across (-90°, 90°) with a high polarization degree exceeding 60%. Our work contributes to the development of structured gamma rays, potentially opening new avenues in high-energy physics, nuclear science, and laboratory astrophysics.

physics.plasm-ph↗

arXiv2025

Measuring the Unspoken: A Disentanglement Model and Benchmark for Psychological Analysis in the Wild

Generative psychological analysis of in-the-wild conversations faces two fundamental challenges: (1) existing Vision-Language Models (VLMs) fail to resolve Articulatory-Affective Ambiguity, where visual patterns of speech mimic emotional expressions; and (2) progress is stifled by a lack of verifiable evaluation metrics capable of assessing visual grounding and reasoning depth. We propose a complete ecosystem to address these twin challenges. First, we introduce Multilevel Insight Network for Disentanglement(MIND), a novel hierarchical visual encoder that introduces a Status Judgment module to algorithmically suppress ambiguous lip features based on their temporal feature variance, achieving explicit visual disentanglement. Second, we construct ConvoInsight-DB, a new large-scale dataset with expert annotations for micro-expressions and deep psychological inference. Third, Third, we designed the Mental Reasoning Insight Rating Metric (PRISM), an automated dimensional framework that uses expert-guided LLM to measure the multidimensional performance of large mental vision models. On our PRISM benchmark, MIND significantly outperforms all baselines, achieving a +86.95% gain in micro-expression detection over prior SOTA. Ablation studies confirm that our Status Judgment disentanglement module is the most critical component for this performance leap. Our code has been opened.

cs.CV↗

arXiv2025

R$^{2}$Seg: Training-Free OOD Medical Tumor Segmentation via Anatomical Reasoning and Statistical Rejection

Foundation models for medical image segmentation struggle under out-of-distribution (OOD) shifts, often producing fragmented false positives on OOD tumors. We introduce R$^{2}$Seg, a training-free framework for robust OOD tumor segmentation that operates via a two-stage Reason-and-Reject process. First, the Reason step employs an LLM-guided anatomical reasoning planner to localize organ anchors and generate multi-scale ROIs. Second, the Reject step applies two-sample statistical testing to candidates generated by a frozen foundation model (BiomedParse) within these ROIs. This statistical rejection filter retains only candidates significantly different from normal tissue, effectively suppressing false positives. Our framework requires no parameter updates, making it compatible with zero-update test-time augmentation and avoiding catastrophic forgetting. On multi-center and multi-modal tumor segmentation benchmarks, R$^{2}$Seg substantially improves Dice, specificity, and sensitivity over strong baselines and the original foundation models. Code are available at https://github.com/Eurekashen/R2Seg.

cs.CV↗

arXiv2025

MELDAE: A Framework for Micro-Expression Spotting, Detection, and Automatic Evaluation in In-the-Wild Conversational Scenes

Accurately analyzing spontaneous, unconscious micro-expressions is crucial for revealing true human emotions, but this task remains challenging in wild scenarios, such as natural conversation. Existing research largely relies on datasets from controlled laboratory environments, and their performance degrades dramatically in the real world. To address this issue, we propose three contributions: the first micro-expression dataset focused on conversational-in-the-wild scenarios; an end-to-end localization and detection framework, MELDAE; and a novel boundary-aware loss function that improves temporal accuracy by penalizing onset and offset errors. Extensive experiments demonstrate that our framework achieves state-of-the-art results on the WDMD dataset, improving the key F1_{DR} localization metric by 17.72% over the strongest baseline, while also demonstrating excellent generalization capabilities on existing benchmarks.

cs.CV↗

arXiv2025

Generative Sequential Notification Optimization via Multi-Objective Decision Transformers

Notifications are an important communication channel for delivering timely and relevant information. Optimizing their delivery involves addressing complex sequential decision-making challenges under constraints such as message utility and user fatigue. Offline reinforcement learning (RL) methods, such as Conservative Q-Learning (CQL), have been applied to this problem but face practical challenges at scale, including instability, sensitivity to distribution shifts, limited reproducibility, and difficulties with explainability in high-dimensional recommendation settings. We present a Decision Transformer (DT) based framework that reframes policy learning as return-conditioned supervised learning, improving robustness, scalability, and modeling flexibility. Our contributions include a real-world comparison with CQL, a multi-reward design suitable for non-episodic tasks, a quantile regression approach to return-to-go conditioning, and a production-ready system with circular buffer-based sequence processing for near-real-time inference. Extensive offline and online experiments in a deployed notification system show that our approach improves notification utility and overall session activity while minimizing user fatigue. Compared to a multi-objective CQL-based agent, the DT-based approach achieved a +0.72% increase in sessions for notification decision-making at LinkedIn by making notification recommendation more relevant.

cs.LG↗

arXiv2025

GOSU: Retrieval-Augmented Generation with Global-Level Optimized Semantic Unit-Centric Framework

Building upon the standard graph-based Retrieval-Augmented Generation (RAG), the introduction of heterogeneous graphs and hypergraphs aims to enrich retrieval and generation by leveraging the relationships between multiple entities through the concept of semantic units (SUs). But this also raises a key issue: The extraction of high-level SUs limited to local text chunks is prone to ambiguity, complex coupling, and increased retrieval overhead due to the lack of global knowledge or the neglect of fine-grained relationships. To address these issues, we propose GOSU, a semantic unit-centric RAG framework that efficiently performs global disambiguation and utilizes SUs to capture interconnections between different nodes across the global context. In the graph construction phase, GOSU performs global merging on the pre-extracted SUs from local text chunks and guides entity and relationship extraction, reducing the difficulty of coreference resolution while uncovering global semantic objects across text chunks. In the retrieval and generation phase, we introduce hierarchical keyword extraction and semantic unit completion. The former uncovers the fine-grained binary relationships overlooked by the latter, while the latter compensates for the coarse-grained n-ary relationships missing from the former. Evaluation across multiple tasks demonstrates that GOSU outperforms the baseline RAG methods in terms of generation quality.

cs.CL↗

arXiv2025

Anyon Theory and Topological Frustration of High-Efficiency Quantum Low-Density Parity-Check Codes

Quantum low-density parity-check (QLDPC) codes offer a promising path to low-overhead fault-tolerant quantum computation but lack systematic strategies for exploration. In this Letter, we establish a topological framework for studying the bivariate-bicycle codes, a prominent class of QLDPC codes tailored for real-world quantum hardware. Our framework enables the investigation of these codes through universal properties of topological orders. In addition to efficient characterizations using Gröbner bases, we also introduce a novel algebraic-geometric approach based on the Bernstein--Khovanskii--Kushnirenko theorem. This approach allows us to analytically determine how the topological order varies with the generic choices of bivariate-bicycle codes under toric layouts. Novel phenomena are unveiled, including topological frustration, where ground-state degeneracy on a torus deviates from the total anyon number, and quasi-fractonic mobility, where anyon movement violates energy conservation. We demonstrate their intrinsic link to symmetry-enriched topological orders and derive an efficient method for generating finite-size codes. Furthermore, we extend the connection between anyons and logical operators using Koszul complex theory. Our Letter provides a rigorous theoretical basis for exploring the fault tolerance of QLDPC codes and deepens the interplay among topological order, quantum error correction, and advanced algebraic structures.

quant-ph↗

arXiv2025

Dynamic Pattern Alignment Learning for Pretraining Lightweight Human-Centric Vision Models

Human-centric vision models (HVMs) have achieved remarkable generalization due to large-scale pretraining on massive person images. However, their dependence on large neural architectures and the restricted accessibility of pretraining data significantly limits their practicality in real-world applications. To address this limitation, we propose Dynamic Pattern Alignment Learning (DPAL), a novel distillation-based pretraining framework that efficiently trains lightweight HVMs to acquire strong generalization from large HVMs. In particular, human-centric visual perception are highly dependent on three typical visual patterns, including global identity pattern, local shape pattern and multi-person interaction pattern. To achieve generalizable lightweight HVMs, we firstly design a dynamic pattern decoder (D-PaDe), acting as a dynamic Mixture of Expert (MoE) model. It incorporates three specialized experts dedicated to adaptively extract typical visual patterns, conditioned on both input image and pattern queries. And then, we present three levels of alignment objectives, which aims to minimize generalization gap between lightweight HVMs and large HVMs at global image level, local pixel level, and instance relation level. With these two deliberate designs, the DPAL effectively guides lightweight model to learn all typical human visual patterns from large HVMs, which can generalize to various human-centric vision tasks. Extensive experiments conducted on 15 challenging datasets demonstrate the effectiveness of the DPAL. Remarkably, when employing PATH-B as the teacher, DPAL-ViT/Ti (5M parameters) achieves surprising generalizability similar to existing large HVMs such as PATH-B (84M) and Sapiens-L (307M), and outperforms previous distillation-based pretraining methods including Proteus-ViT/Ti (5M) and TinyMiM-ViT/Ti (5M) by a large margin.

cs.CV↗

arXiv2025

AI PsyRoom: Artificial Intelligence Platform for Segmented Yearning and Reactive Outcome Optimization Method

Psychological counseling faces huge challenges due to the growing demand for mental health services and the shortage of trained professionals. Large language models (LLMs) have shown potential to assist psychological counseling, especially in empathy and emotional support. However, existing models lack a deep understanding of emotions and are unable to generate personalized treatment plans based on fine-grained emotions. To address these shortcomings, we present AI PsyRoom, a multi-agent simulation framework designed to enhance psychological counseling by generating empathetic and emotionally nuanced conversations. By leveraging fine-grained emotion classification and a multi-agent framework, we construct a multi-agent PsyRoom A for dialogue reconstruction, generating a high-quality dialogue dataset EmoPsy, which contains 35 sub-emotions, 423 specific emotion scenarios, and 12,350 dialogues. We also propose PsyRoom B for generating personalized treatment plans. Quantitative evaluations demonstrate that AI PsyRoom significantly outperforms state-of-the-art methods, achieving 18% improvement in problem orientation, 23% in expression, 24% in Empathy, and 16% in interactive communication quality. The datasets and models are publicly available, providing a foundation for advancing AI-assisted psychological counseling research.

cs.AI↗

arXiv2025

From Sentences to Sequences: Rethinking Languages in Biological System

The paradigm of large language models in natural language processing (NLP) has also shown promise in modeling biological languages, including proteins, RNA, and DNA. Both the auto-regressive generation paradigm and evaluation metrics have been transferred from NLP to biological sequence modeling. However, the intrinsic structural correlations in natural and biological languages differ fundamentally. Therefore, we revisit the notion of language in biological systems to better understand how NLP successes can be effectively translated to biological domains. By treating the 3D structure of biomolecules as the semantic content of a sentence and accounting for the strong correlations between residues or bases, we highlight the importance of structural evaluation and demonstrate the applicability of the auto-regressive paradigm in biological language modeling. Code can be found at \href{https://github.com/zjuKeLiu/RiFold}{github.com/zjuKeLiu/RiFold}

q-bio.BM↗

arXiv2025

Generalized toric codes on twisted tori for quantum error correction

The Kitaev toric code is widely considered one of the leading candidates for error correction in fault-tolerant quantum computation. However, direct methods to increase its logical dimensions, such as lattice surgery or introducing punctures, often incur prohibitive overheads. In this work, we introduce a ring-theoretic approach for efficiently analyzing topological CSS codes in two dimensions, enabling the exploration of generalized toric codes with larger logical dimensions on twisted tori. Using Gröbner bases, we simplify stabilizer syndromes to efficiently identify anyon excitations and their geometric periodicities, even under twisted periodic boundary conditions. Since the properties of the codes are determined by the anyons, this approach allows us to directly compute the logical dimensions without constructing large parity-check matrices. Our approach provides a unified method for finding new quantum error-correcting codes and exhibiting their underlying topological orders via the Laurent polynomial ring. This framework naturally applies to bivariate bicycle codes. For example, we construct optimal weight-6 generalized toric codes on twisted tori with parameters $[[ n, k, d ]]$ for $n \leq 400$, yielding novel codes such as $[[120,8,12]]$, $[[186,10,14]]$, $[[210,10,16]]$, $[[248, 10, 18]]$, $[[254, 14, 16]]$, $[[294, 10, 20]]$, $[[310, 10, \leq 22]]$, and $[[340, 16, 18]]$. Moreover, we present a new realization of the $[[360, 12, \leq 24]]$ quantum code using the $(3,3)$-bivariate bicycle code on a twisted torus defined by the basis vectors $(0,30)$ and $(6,6)$, improving stabilizer locality relative to the previous construction. These results highlight the power of the topological order perspective in advancing the design and theoretical understanding of quantum low-density parity-check (LDPC) codes.

quant-ph↗

arXiv2025

Less is More: DocString Compression in Code Generation

The widespread use of Large Language Models (LLMs) in software engineering has intensified the need for improved model and resource efficiency. In particular, for neural code generation, LLMs are used to translate function/method signature and DocString to executable code. DocStrings which capture user re quirements for the code and used as the prompt for LLMs, often contains redundant information. Recent advancements in prompt compression have shown promising results in Natural Language Processing (NLP), but their applicability to code generation remains uncertain. Our empirical study show that the state-of-the-art prompt compression methods achieve only about 10% reduction, as further reductions would cause significant performance degradation. In our study, we propose a novel compression method, ShortenDoc, dedicated to DocString compression for code generation. Our extensive experiments on six code generation datasets, five open-source LLMs (1B to 10B parameters), and one closed-source LLM GPT-4o confirm that ShortenDoc achieves 25-40% compression while preserving the quality of generated code, outperforming other baseline methods at similar compression levels. The benefit of this research is to improve efficiency and reduce the cost while maintaining the quality of the generated code, especially when calling third-party APIs, and is able to reduce the token processing cost by 25-40%.

cs.SE↗

arXiv2025

Human-machine collaboration: ordering mechanism of rank-2 spin liquid on breathing pyrochlore lattice

Machine learning algorithms thrive on large data sets of good quality. Here we show that they can also excel in a typical research setting with little data of limited quality, through an interplay of insights coming from machine, and human researchers. The question we address is the unsolved problem of ordering out of a spin-liquid phase described by an emergent rank-2 U(1) gauge theory, as described by [H. Yan et al., Phys. Rev. Lett. 124, 127203 (2020)]. Published Monte Carlo simulations for this problem are consistent with a strong first-order phase transition, but were too noisy for the form of low-temperature order to be identified. Using a highly-interpretable machine learning approach based on a support vector machine with a tensorial kernel (TKSVM), we re-analyze this Monte Carlo data, gaining new information about the form of order that could in turn be interpreted by traditionally-trained physicists. We find that the low-temperature ordered phase is a form of magnetic order analogous to a smectic liquid crystal. This arises due to a subtle thermal order-by-disorder mechanism, that can be understood from the fluctuations of the tensor electric field of the parent rank-2 gauge theory. These results were obtained by a back-and-forth process which closely resembles a collaboration between human researchers and machines. We argue that this "collaborative" approach may provide a blueprint for solving other problems that have not yielded to human insights alone.

cond-mat.str-el↗

arXiv2025

ImputeINR: Time Series Imputation via Implicit Neural Representations for Disease Diagnosis with Missing Data

Healthcare data frequently contain a substantial proportion of missing values, necessitating effective time series imputation to support downstream disease diagnosis tasks. However, existing imputation methods focus on discrete data points and are unable to effectively model sparse data, resulting in particularly poor performance for imputing substantial missing values. In this paper, we propose a novel approach, ImputeINR, for time series imputation by employing implicit neural representations (INR) to learn continuous functions for time series. ImputeINR leverages the merits of INR in that the continuous functions are not coupled to sampling frequency and have infinite sampling frequency, allowing ImputeINR to generate fine-grained imputations even on extremely sparse observed values. Extensive experiments conducted on eight datasets with five ratios of masked values show the superior imputation performance of ImputeINR, especially for high missing ratios in time series data. Furthermore, we validate that applying ImputeINR to impute missing values in healthcare data enhances the performance of downstream disease diagnosis tasks. Codes are available.

cs.LG↗

arXiv2025

TSINR: Capturing Temporal Continuity via Implicit Neural Representations for Time Series Anomaly Detection

Time series anomaly detection aims to identify unusual patterns in data or deviations from systems' expected behavior. The reconstruction-based methods are the mainstream in this task, which learn point-wise representation via unsupervised learning. However, the unlabeled anomaly points in training data may cause these reconstruction-based methods to learn and reconstruct anomalous data, resulting in the challenge of capturing normal patterns. In this paper, we propose a time series anomaly detection method based on implicit neural representation (INR) reconstruction, named TSINR, to address this challenge. Due to the property of spectral bias, TSINR enables prioritizing low-frequency signals and exhibiting poorer performance on high-frequency abnormal data. Specifically, we adopt INR to parameterize time series data as a continuous function and employ a transformer-based architecture to predict the INR of given data. As a result, the proposed TSINR method achieves the advantage of capturing the temporal continuity and thus is more sensitive to discontinuous anomaly data. In addition, we further design a novel form of INR continuous function to learn inter- and intra-channel information, and leverage a pre-trained large language model to amplify the intense fluctuations in anomalies. Extensive experiments demonstrate that TSINR achieves superior overall performance on both univariate and multivariate time series anomaly detection benchmarks compared to other state-of-the-art reconstruction-based methods. Our codes are available.

cs.LG↗

arXiv2025

Uncertainty-Aware Multi-Expert Knowledge Distillation for Imbalanced Disease Grading

Automatic disease image grading is a significant application of artificial intelligence for healthcare, enabling faster and more accurate patient assessments. However, domain shifts, which are exacerbated by data imbalance, introduce bias into the model, posing deployment difficulties in clinical applications. To address the problem, we propose a novel \textbf{U}ncertainty-aware \textbf{M}ulti-experts \textbf{K}nowledge \textbf{D}istillation (UMKD) framework to transfer knowledge from multiple expert models to a single student model. Specifically, to extract discriminative features, UMKD decouples task-agnostic and task-specific features with shallow and compact feature alignment in the feature space. At the output space, an uncertainty-aware decoupled distillation (UDD) mechanism dynamically adjusts knowledge transfer weights based on expert model uncertainties, ensuring robust and reliable distillation. Additionally, UMKD also tackles the problems of model architecture heterogeneity and distribution discrepancies between source and target domains, which are inadequately tackled by previous KD approaches. Extensive experiments on histology prostate grading (\textit{SICAPv2}) and fundus image grading (\textit{APTOS}) demonstrate that UMKD achieves a new state-of-the-art in both source-imbalanced and target-imbalanced scenarios, offering a robust and practical solution for real-world disease image grading.

cs.CV↗

arXiv2025

A Dynamic Fuzzy Rule and Attribute Management Framework for Fuzzy Inference Systems in High-Dimensional Data

This paper presents an Adaptive Dynamic Attribute and Rule (ADAR) framework designed to address the challenges posed by high-dimensional data in neuro-fuzzy inference systems. By integrating dual weighting mechanisms-assigning adaptive importance to both attributes and rules-together with automated growth and pruning strategies, ADAR adaptively streamlines complex fuzzy models without sacrificing performance or interpretability. Experimental evaluations on four diverse datasets - Auto MPG (7 variables), Beijing PM2.5 (10 variables), Boston Housing (13 variables), and Appliances Energy Consumption (27 variables) show that ADAR-based models achieve consistently lower Root Mean Square Error (RMSE) compared to state-of-the-art baselines. On the Beijing PM2.5 dataset, for instance, ADAR-SOFENN attained an RMSE of 56.87 with nine rules, surpassing traditional ANFIS [12] and SOFENN [16] models. Similarly, on the high-dimensional Appliances Energy dataset, ADAR-ANFIS reached an RMSE of 83.25 with nine rules, outperforming established fuzzy logic approaches and interpretability-focused methods such as APLR. Ablation studies further reveal that combining rule-level and attribute-level weight assignment significantly reduces model overlap while preserving essential features, thereby enhancing explainability. These results highlight ADAR's effectiveness in dynamically balancing rule complexity and feature importance, paving the way for scalable, high-accuracy, and transparent neuro-fuzzy systems applicable to a range of real-world scenarios.

cs.AI↗