arXiv subjects

Jun Xu

Publications and source records attributed to Jun Xu.

At least 109 records · Page 6Linked to original sources

arXiv2025

Thinker: Training LLMs in Hierarchical Thinking for Deep Search via Multi-Turn Interaction

Efficient retrieval of external knowledge bases and web pages is crucial for enhancing the reasoning abilities of LLMs. Previous works on training LLMs to leverage external retrievers for solving complex problems have predominantly employed end-to-end reinforcement learning. However, these approaches neglect supervision over the reasoning process, making it difficult to guarantee logical coherence and rigor. To address these limitations, we propose Thinker, a hierarchical thinking model for deep search through multi-turn interaction, making the reasoning process supervisable and verifiable. It decomposes complex problems into independently solvable sub-problems, each dually represented in both natural language and an equivalent logical function to support knowledge base and web searches. Concurrently, dependencies between sub-problems are passed as parameters via these logical functions, enhancing the logical coherence of the problem-solving process. To avoid unnecessary external searches, we perform knowledge boundary determination to check if a sub-problem is within the LLM's intrinsic knowledge, allowing it to answer directly. Experimental results indicate that with as few as several hundred training samples, the performance of Thinker is competitive with established baselines. Furthermore, when scaled to the full training set, Thinker significantly outperforms these methods across various datasets and model sizes. The source code is available at https://github.com/OpenSPG/KAG-Thinker.

cs.AI↗

arXiv2025

Rethinking the Evaluation of Secure Code Generation

Large language models (LLMs) are widely used in software development. However, the code generated by LLMs often contains vulnerabilities. Several secure code generation methods have been proposed to address this issue, but their current evaluation schemes leave several concerns unaddressed. Specifically, most existing studies evaluate security and functional correctness separately, using different datasets. That is, they assess vulnerabilities using security-related code datasets while validating functionality with general code datasets. In addition, prior research primarily relies on a single static analyzer, CodeQL, to detect vulnerabilities in generated code, which limits the scope of security evaluation. In this work, we conduct a comprehensive study to systematically assess the improvements introduced by four state-of-the-art secure code generation techniques. Specifically, we apply both security inspection and functionality validation to the same generated code and evaluate these two aspects together. We also employ three popular static analyzers and two LLMs to identify potential vulnerabilities in the generated code. Our study reveals that existing techniques often compromise the functionality of generated code to enhance security. Their overall performance remains limited when evaluating security and functionality together. In fact, many techniques even degrade the performance of the base LLM by more than 50%. Our further inspection reveals that these techniques often either remove vulnerable lines of code entirely or generate ``garbage code'' that is unrelated to the intended task. Moreover, the commonly used static analyzer CodeQL fails to detect several vulnerabilities, further obscuring the actual security improvements achieved by existing techniques.

cs.CR↗

arXiv2025

TraceCoder: Towards Traceable ICD Coding via Multi-Source Knowledge Integration

Automated International Classification of Diseases (ICD) coding assigns standardized diagnosis and procedure codes to clinical records, playing a critical role in healthcare systems. However, existing methods face challenges such as semantic gaps between clinical text and ICD codes, poor performance on rare and long-tail codes, and limited interpretability. To address these issues, we propose TraceCoder, a novel framework integrating multi-source external knowledge to enhance traceability and explainability in ICD coding. TraceCoder dynamically incorporates diverse knowledge sources, including UMLS, Wikipedia, and large language models (LLMs), to enrich code representations, bridge semantic gaps, and handle rare and ambiguous codes. It also introduces a hybrid attention mechanism to model interactions among labels, clinical context, and knowledge, improving long-tail code recognition and making predictions interpretable by grounding them in external evidence. Experiments on MIMIC-III-ICD9, MIMIC-IV-ICD9, and MIMIC-IV-ICD10 datasets demonstrate that TraceCoder achieves state-of-the-art performance, with ablation studies validating the effectiveness of its components. TraceCoder offers a scalable and robust solution for automated ICD coding, aligning with clinical needs for accuracy, interpretability, and reliability.

cs.CL↗

arXiv2025

TACL: Threshold-Adaptive Curriculum Learning Strategy for Enhancing Medical Text Understanding

Medical texts, particularly electronic medical records (EMRs), are a cornerstone of modern healthcare, capturing critical information about patient care, diagnoses, and treatments. These texts hold immense potential for advancing clinical decision-making and healthcare analytics. However, their unstructured nature, domain-specific language, and variability across contexts make automated understanding an intricate challenge. Despite the advancements in natural language processing, existing methods often treat all data as equally challenging, ignoring the inherent differences in complexity across clinical records. This oversight limits the ability of models to effectively generalize and perform well on rare or complex cases. In this paper, we present TACL (Threshold-Adaptive Curriculum Learning), a novel framework designed to address these challenges by rethinking how models interact with medical texts during training. Inspired by the principle of progressive learning, TACL dynamically adjusts the training process based on the complexity of individual samples. By categorizing data into difficulty levels and prioritizing simpler cases early in training, the model builds a strong foundation before tackling more complex records. By applying TACL to multilingual medical data, including English and Chinese clinical records, we observe significant improvements across diverse clinical tasks, including automatic ICD coding, readmission prediction and TCM syndrome differentiation. TACL not only enhances the performance of automated systems but also demonstrates the potential to unify approaches across disparate medical domains, paving the way for more accurate, scalable, and globally applicable medical text understanding solutions.

cs.CL↗

arXiv2025

LLaDA-Rec: Discrete Diffusion for Parallel Semantic ID Generation in Generative Recommendation

Generative recommendation represents each item as a semantic ID, i.e., a sequence of discrete tokens, and generates the next item through autoregressive decoding. While effective, existing autoregressive models face two intrinsic limitations: (1) unidirectional constraints, where causal attention restricts each token to attend only to its predecessors, hindering global semantic modeling; and (2) error accumulation, where the fixed left-to-right generation order causes prediction errors in early tokens to propagate to the predictions of subsequent token. To address these issues, we propose LLaDA-Rec, a discrete diffusion framework that reformulates recommendation as parallel semantic ID generation. By combining bidirectional attention with the adaptive generation order, the approach models inter-item and intra-item dependencies more effectively and alleviates error accumulation. Specifically, our approach comprises three key designs: (1) a parallel tokenization scheme that produces semantic IDs for bidirectional modeling, addressing the mismatch between residual quantization and bidirectional architectures; (2) two masking mechanisms at the user-history and next-item levels to capture both inter-item sequential dependencies and intra-item semantic relationships; and (3) an adapted beam search strategy for adaptive-order discrete diffusion decoding, resolving the incompatibility of standard beam search with diffusion-based generation. Experiments on three real-world datasets show that LLaDA-Rec consistently outperforms both ID-based and state-of-the-art generative recommenders, establishing discrete diffusion as a new paradigm for generative recommendation.

cs.IR↗

arXiv2025

Introducing LongCat-Flash-Thinking: A Technical Report

We present LongCat-Flash-Thinking, an efficient 560-billion-parameter open-source Mixture-of-Experts (MoE) reasoning model. Its advanced capabilities are cultivated through a meticulously crafted training process, beginning with long Chain-of-Thought (CoT) data cold-start and culminating in large-scale Reinforcement Learning (RL). We first employ a well-designed cold-start training strategy, which significantly enhances the reasoning potential and equips the model with specialized skills in both formal and agentic reasoning. Then, a core innovation is our domain-parallel training scheme, which decouples optimization across distinct domains (e.g., STEM, Code, Agentic) and subsequently fuses the resulting expert models into a single, nearly Pareto-optimal model. This entire process is powered by our Dynamic ORchestration for Asynchronous rollout (DORA) system, a large-scale RL framework that delivers a greater than threefold training speedup over synchronous methods on tens of thousands of accelerators. As a result, LongCat-Flash-Thinking achieves state-of-the-art performance among open-source models on a suite of complex reasoning tasks. The model exhibits exceptional efficiency in agentic reasoning, reducing average token consumption by 64.5% (from 19, 653 to 6, 965) on AIME-25, without degrading task accuracy. We release LongCat-Flash-Thinking to promote further advances in reasoning systems and agentic AI research.

cs.AI↗

arXiv2025

Every Activation Boosted: Scaling General Reasoner to 1 Trillion Open Language Foundation

We introduce Ling 2.0, a series reasoning-oriented language foundation built upon the principle that every activation boosts reasoning capability. Designed to scale from tens of billions to one trillion parameters under a unified Mixture-of-Experts (MoE) paradigm, Ling 2.0 emphasizes high sparsity, cross-scale consistency, and efficiency guided by empirical scaling laws. The series includes three non-thinking (instruct) models - Ling-mini-2.0, Ling-flash-2.0, and Ling-1T - ranging from 16B to 1T total parameters and achieving up to 7-fold active-compute efficiency compared with dense counterparts. Ling 2.0 integrates coordinated innovations across model architecture, pre-training, post-training, and infrastructure: a high-sparsity MoE with MTP for efficient reasoning, reasoning-oriented data and mid-training CoT activation, reinforcement-based fine-tuning (DFT, Evo-CoT), and full-scale FP8 training with fine-grained heterogeneous pipelines. At the trillion scale, Ling-1T establishes a new Pareto frontier of reasoning accuracy versus computational efficiency, demonstrating that sparse activation, when properly aligned with reasoning objectives, enables scalable and efficient intelligence. Collectively, Ling 2.0 provides a coherent, open, and efficient foundation for advancing future reasoning and thinking models, including the Ring series built upon the same base.

cs.CL↗

arXiv2025

Learning the PTM Code through a Coarse-to-Fine, Mechanism-Aware Framework

Post-translational modifications (PTMs) form a combinatorial "code" that regulates protein function, yet deciphering this code - linking modified sites to their catalytic enzymes - remains a central unsolved problem in understanding cellular signaling and disease. We introduce COMPASS-PTM, a mechanism-aware, coarse-to-fine learning framework that unifies residue-level PTM profiling with enzyme-substrate assignment. COMPASS-PTM integrates evolutionary representations from protein language models with physicochemical priors and a crosstalk-aware prompting mechanism that explicitly models inter-PTM dependencies. This design allows the model to learn biologically coherent patterns of cooperative and antagonistic modifications while addressing the dual long-tail distribution of PTM data. Across multiple proteome-scale benchmarks, COMPASS-PTM establishes new state-of-the-art performance, including a 122% relative F1 improvement in multi-label site prediction and a 54% gain in zero-shot enzyme assignment. Beyond accuracy, the model demonstrates interpretable generalization, recovering canonical kinase motifs and predicting disease-associated PTM rewiring caused by missense variants. By bridging statistical learning with biochemical mechanism, COMPASS-PTM unifies site-level and enzyme-level prediction into a single framework that learns the grammar underlying protein regulation and signaling.

cs.CE↗

arXiv2025

CommonSense: Efficient Set Intersection (SetX) Protocol Based on Compressed Sensing

In the set reconciliation (\textsf{SetR}) problem, two parties Alice and Bob, holding sets $\mathsf{A}$ and $\mathsf{B}$, communicate to learn the symmetric difference $\mathsf{A} Δ\mathsf{B}$. In this work, we study a related but under-explored problem: set intersection (\textsf{SetX})~\cite{Ozisik2019}, where both parties learn $\mathsf{A} \cap \mathsf{B}$ instead. However, existing solutions typically reuse \textsf{SetR} protocols due to the absence of dedicated \textsf{SetX} protocols and the misconception that \textsf{SetR} and \textsf{SetX} have comparable costs. Observing that \textsf{SetX} is fundamentally cheaper than \textsf{SetR}, we developed a multi-round \textsf{SetX} protocol that outperforms the information-theoretic lower bound of \textsf{SetR} problem. In our \textsf{SetX} protocol, Alice sends Bob a compressed sensing (CS) sketch of $\mathsf{A}$ to help Bob identify his unique elements (those in $\mathsf{B \setminus A}$). This solves the \textsf{SetX} problem, if $\mathsf{A} \subseteq \mathsf{B}$. Otherwise, Bob sends a CS sketch of the residue (a set of elements he cannot decode) back to Alice for her to decode her unique elements (those in $\mathsf{A \setminus B}$). As such, Alice and Bob communicate back and forth %with a set membership filter (SMF) of estimated $\mathsf{B \setminus A}$. Alice updates $\mathsf{A}$ and communication repeats until both parties agrees on $\mathsf{A} \cap \mathsf{B}$. On real world datasets, experiments show that our $\mathsf{SetX}$ protocol reduces the communication cost by 8 to 10 times compared to the IBLT-based $\mathsf{SetR}$ protocol.

cs.DC↗

arXiv2025

Merge and Guide: Unifying Model Merging and Guided Decoding for Controllable Multi-Objective Generation

Adapting to diverse user needs at test time is a key challenge in controllable multi-objective generation. Existing methods are insufficient: merging-based approaches provide indirect, suboptimal control at the parameter level, often disregarding the impacts of multiple objectives. While decoding-based guidance is more direct, it typically requires aggregating logits from multiple expert models, incurring significant space overhead and relying heavily on individual model capacity. To address these issues, we introduce Merge-And-GuidE (MAGE), a two-stage framework that leverages model merging for guided decoding. We first identify a critical compatibility problem between the guidance and base models. In Stage 1, MAGE resolves this by dynamically constructing a more robust base model, merging a series of backbone models that account for multiple objectives. In Stage 2, we merge explicit and implicit value models into a unified guidance proxy, which then steers the decoding of the base model from Stage 1. Our analysis empirically validates Linear Mode Connectivity (LMC) in value models, explores the relationship between model merging and prediction ensembling, and demonstrates the enhanced controllability afforded by our approach. Extensive experiments show that our method outperforms existing approaches, achieving superior controllability, Pareto-optimal performance, and enhanced adaptability.

cs.LG↗

arXiv2025

A Hierarchical Geometry-guided Transformer for Histological Subtyping of Primary Liver Cancer

Primary liver malignancies are widely recognized as the most heterogeneous and prognostically diverse cancers of the digestive system. Among these, hepatocellular carcinoma (HCC) and intrahepatic cholangiocarcinoma (ICC) emerge as the two principal histological subtypes, demonstrating significantly greater complexity in tissue morphology and cellular architecture than other common tumors. The intricate representation of features in Whole Slide Images (WSIs) encompasses abundant crucial information for liver cancer histological subtyping, regarding hierarchical pyramid structure, tumor microenvironment (TME), and geometric representation. However, recent approaches have not adequately exploited these indispensable effective descriptors, resulting in a limited understanding of histological representation and suboptimal subtyping performance. To mitigate these limitations, ARGUS is proposed to advance histological subtyping in liver cancer by capturing the macro-meso-micro hierarchical information within the TME. Specifically, we first construct a micro-geometry feature to represent fine-grained cell-level pattern via a geometric structure across nuclei, thereby providing a more refined and precise perspective for delineating pathological images. Then, a Hierarchical Field-of-Views (FoVs) Alignment module is designed to model macro- and meso-level hierarchical interactions inherent in WSIs. Finally, the augmented micro-geometry and FoVs features are fused into a joint representation via present Geometry Prior Guided Fusion strategy for modeling holistic phenotype interactions. Extensive experiments on public and private cohorts demonstrate that our ARGUS achieves state-of-the-art (SOTA) performance in histological subtyping of liver cancer, which provide an effective diagnostic tool for primary liver malignancies in clinical practice.

cs.CV↗

arXiv2025

A Self-Adaptive Frequency Domain Network for Continuous Intraoperative Hypotension Prediction

Intraoperative hypotension (IOH) is strongly associated with postoperative complications, including postoperative delirium and increased mortality, making its early prediction crucial in perioperative care. While several artificial intelligence-based models have been developed to provide IOH warnings, existing methods face limitations in incorporating both time and frequency domain information, capturing short- and long-term dependencies, and handling noise sensitivity in biosignal data. To address these challenges, we propose a novel Self-Adaptive Frequency Domain Network (SAFDNet). Specifically, SAFDNet integrates an adaptive spectral block, which leverages Fourier analysis to extract frequency-domain features and employs self-adaptive thresholding to mitigate noise. Additionally, an interactive attention block is introduced to capture both long-term and short-term dependencies in the data. Extensive internal and external validations on two large-scale real-world datasets demonstrate that SAFDNet achieves up to 97.3\% AUROC in IOH early warning, outperforming state-of-the-art models. Furthermore, SAFDNet exhibits robust predictive performance and low sensitivity to noise, making it well-suited for practical clinical applications.

cs.LG↗

arXiv2025

OnePiece: Bringing Context Engineering and Reasoning to Industrial Cascade Ranking System

Despite the growing interest in replicating the scaled success of large language models (LLMs) in industrial search and recommender systems, most existing industrial efforts remain limited to transplanting Transformer architectures, which bring only incremental improvements over strong Deep Learning Recommendation Models (DLRMs). From a first principle perspective, the breakthroughs of LLMs stem not only from their architectures but also from two complementary mechanisms: context engineering, which enriches raw input queries with contextual cues to better elicit model capabilities, and multi-step reasoning, which iteratively refines model outputs through intermediate reasoning paths. However, these two mechanisms and their potential to unlock substantial improvements remain largely underexplored in industrial ranking systems. In this paper, we propose OnePiece, a unified framework that seamlessly integrates LLM-style context engineering and reasoning into both retrieval and ranking models of industrial cascaded pipelines. OnePiece is built on a pure Transformer backbone and further introduces three key innovations: (1) structured context engineering, which augments interaction history with preference and scenario signals and unifies them into a structured tokenized input sequence for both retrieval and ranking; (2) block-wise latent reasoning, which equips the model with multi-step refinement of representations and scales reasoning bandwidth via block size; (3) progressive multi-task training, which leverages user feedback chains to effectively supervise reasoning steps during training. OnePiece has been deployed in the main personalized search scenario of Shopee and achieves consistent online gains across different key business metrics, including over $+2\%$ GMV/UU and a $+2.90\%$ increase in advertising revenue.

cs.IR↗

arXiv2025

Global Spin Alignment of (Anti-)$^4$Li in Non-Central Heavy-Ion Collisions

Non-central heavy-ion collisions produce hot and dense nuclear matter with significant fluid vorticity, which can induce global polarizations or alignments of particles with non-zero spins along the direction of the total orbital angular momentum. This phenomenon has been observed for hyperons and vector mesons in experiments. In the present study, we demonstrate that polarized nucleons lead to global spin alignment of the unstable nucleus $^4$Li, which can be measured through its strong decays via $^4\text{Li} \rightarrow {^3\text{He}} + p$. Assuming that $^4$Li is formed through the coalescence of polarized nucleons at kinetic freeze-out, we obtain the angular distribution of the daughter particle $^3$He in the rest frame of the polarized $^4$Li. Taking kinetically freeze-out nucleons from an isotropic and thermalized fireball of constant vorticity and including quantum corrections up to $\hbar^2$ in the coalescence calculation through the Moyal star product, we find that the angular distribution of $^3$He has a $\cos(2θ^*)$ dependence with $θ^*$ being its angle with respect to the quantization axis of $^4$Li. We also find that the $^3$He angular distribution depends on both the vorticity and the polarization of kinetically freeze-out nucleons. Future measurements on the spin alignment of $^4$Li in heavy-ion collisions thus offer a promising method to probe the spin dynamics, vortical structure, and spin-dependent equation-of-state of the nuclear matter produced in these collisions.

nucl-th↗

arXiv2025

A comparison study of collisions at relativistic energies involving light nuclei

We present extensive comparisons of $^{16}$O+$^{16}$O collisions at the center-of-mass energy per nucleon pair $\sqrt{s_{NN}}=200$ GeV and $^{208}$Pb+$^{16}$O collisions at $\sqrt{s_{NN}}=68.5$ GeV as well as $^{20}$Ne+$^{20}$Ne collisions at $\sqrt{s_{NN}}=200$ GeV and $^{208}$Pb+$^{20}$Ne collisions at $\sqrt{s_{NN}}=68.5$ GeV based on a multiphase transport (AMPT) model. We recommend measuring the ratio of the elliptic flow to the triangular flow, which shows appreciable sensitivity to the structure of light nuclei as also found in other studies. This is especially so if the observable is measured near the target rapidity in $^{208}$Pb+$^{16}$O or $^{208}$Pb+$^{20}$Ne collisions, as originally found in the present study. Our study serves as a useful reference for understanding the structure effect on observables in collisions involving light nuclei under analysis or on the schedule.

nucl-th↗

arXiv2025

OddEEC: A New Sketch Technique for Error Estimating Coding

Error estimating coding (EEC) is a standard technique for estimating the number of bit errors during packet transmission over wireless networks. In this paper, we propose OddEEC, a novel EEC scheme. OddEEC is a nontrivial adaptation of a data sketching technique named Odd Sketch to EEC, addressing new challenges therein by its bit sampling technique and maximum likelihood estimator. Our experiments show that OddEEC overall achieves comparable estimation accuracy as competing schemes such as gEEC and mEEC, with much smaller decoding complexity.

cs.NI↗

arXiv2025

HealthiVert-GAN: A Novel Framework of Pseudo-Healthy Vertebral Image Synthesis for Interpretable Compression Fracture Grading

Osteoporotic vertebral compression fractures (OVCFs) are prevalent in the elderly population, typically assessed on computed tomography (CT) scans by evaluating vertebral height loss. This assessment helps determine the fracture's impact on spinal stability and the need for surgical intervention. However, the absence of pre-fracture CT scans and standardized vertebral references leads to measurement errors and inter-observer variability, while irregular compression patterns further challenge the precise grading of fracture severity. While deep learning methods have shown promise in aiding OVCFs screening, they often lack interpretability and sufficient sensitivity, limiting their clinical applicability. To address these challenges, we introduce a novel vertebra synthesis-height loss quantification-OVCFs grading framework. Our proposed model, HealthiVert-GAN, utilizes a coarse-to-fine synthesis network designed to generate pseudo-healthy vertebral images that simulate the pre-fracture state of fractured vertebrae. This model integrates three auxiliary modules that leverage the morphology and height information of adjacent healthy vertebrae to ensure anatomical consistency. Additionally, we introduce the Relative Height Loss of Vertebrae (RHLV) as a quantification metric, which divides each vertebra into three sections to measure height loss between pre-fracture and post-fracture states, followed by fracture severity classification using a Support Vector Machine (SVM). Our approach achieves state-of-the-art classification performance on both the Verse2019 dataset and in-house dataset, and it provides cross-sectional distribution maps of vertebral height loss. This practical tool enhances diagnostic accuracy in clinical settings and assisting in surgical decision-making.

eess.IV↗

arXiv2025

Extracting temperature of plasma through fusion reactions within a transport approach

We have investigated the method of extracting the temperature from weighted proton-to-neutron yield ratio from fusion reactions as in the previous experiment~[W. Bang, {\it et al.}, Phys. Rev. Lett. \textbf{111}, 055002 (2013).] using the Texas Petawatt laser beam. The Coulomb explosion of deuterium clusters is simulated based on the particle-in-cell model in a box system with periodic boundary conditions, and fusion reactions are incorporated through the stochastic method. As long as the deuteron numbers in deuterium clusters follow a log-normal distribution, the low-energy part of the final deuteron spectrum can be fitted by a Maxwell-Boltzmann distribution, while there are more deuterons in the intermediate- and high-energy region compared to a thermal distribution, and dominate the weighted yield ratio. Therefore, the effective temperature extracted from the weighted yield ratio is generally higher than that from fitting the final deuteron spectrum. The local density fluctuation, which intrinsically exists due to the log-normal distribution of deuteron numbers, further enhances hot deuteron-deuteron collisions, and significantly affects the weighted yield ratio.

physics.plasm-ph↗