arXiv subjects

Lixin Zhang

Publications and source records attributed to Lixin Zhang.

At least 19 recordsLinked to original sources

arXiv2026

$S^3$-Bench: Evaluating Speech Interaction Models as Scientific Voice Assistants

The advance of multimodal large language models (MLLMs) has fundamentally reshaped the paradigm of human-computer interaction, especially speech interaction models capable of seamless conversations. Despite remarkable performance as general voice assistants, their performance in specialized domains remains underexplored, particularly in scientific areas. Scientific interactions introduce formidable challenges, involving rare technical terminology, spoken norms of abbreviations, and the natural verbalization of symbolic special expressions. In this paper, we introduce S$^3$-Bench, a systematic evaluation framework covering 10 major disciplines, consisting of a Knowledge set for speech question-answering and a Dialogue set for multi-turn progressive interactions with simulated user agents. By decomposing a complete atomic turn into stages of speech recognition, perception, knowledge utilization with reasoning, and response pronunciation, we systematically characterize the common challenges and performance tradeoffs of existing approaches. Furthermore, experiments on multi-turn interactions reveal persistent limitations in user adaptation and the generation of accurate, comprehensive, and efficient responses.

cs.CL↗

arXiv2026

OmniJudge or OmniBias? Diagnosing Multimodal Judges through Balanced, Decoupled Lenses

Multimodal understanding models that can jointly judge text-to-image (T2I), text-to-video (T2V) and text-to-speech (TTS) generation are increasingly used as "OmniJudges" for evaluation and automatic annotation. How reliably they understand what they score remains unclear, since existing benchmarks and training data tend to overemphasize positive examples and to conflate distinct failure modes, so a judge may score well without recognizing failures while its capability gaps stay hidden. Motivated by this, we introduce D3-Omni, a balanced and decoupled benchmark for diagnosing fine-grained multimodal understanding, covering 53 orthogonal binary dimensions (17/22/14) and 10,671 samples (3,526/1,998/5,147) across the three tasks. Rather than re-generating outputs, which may leak information across dimensions, we fix verified fully positive seeds and derive negatives through controlled prompt rewriting and atomic, dimension-isolating perturbations. The resulting D3 design is Dual-balanced, which helps alleviate negative-sample scarcity and per-dimension label imbalance; Decoupled, so that each error is attributable to a single capability; and Dynamic, steering construction toward under-represented regions of the label distribution as generative models improve.The suite reaches near 1:1 per-dimension parity and a uniform distribution over all total-score levels. Under this balanced view, even strong OmniJudges tend to struggle on modality-related dimensions, to confirm satisfied requirements far more reliably than they detect violated ones, and to treat nominally distinct attributes as largely a single decision, suggesting that aggregate accuracy may hide systematic blind spots that a balanced and decoupled lens can help expose and, in turn, address.

cs.AI↗

arXiv2026

Continuity of measure-theoretic entropy for stochastic differential equations

For stochastic differential equations, we establish a relationship between the measure-theoretic entropy of the stochastic flow and the rate of volume growth of stable submanifolds under iteration. By combining the result of Kifer and Yomdin (1988), we show that under a suitable integrability condition, for systems with $C^{\infty}$ coefficients, the measure-theoretic entropy is upper semicontinuous with respect to the coefficients of SDEs.

math.DS↗

arXiv2026

Symmetry enforced quantum spin Hall effect in Altermagnets

The quantum spin Hall effect (QSHE) has attracted widespread attention due to its dissipationless transport, which is protected by non-trivial topological invariants and helical edge states. Because even weak magnetic disorder can destroy the stability of topological quantum states, current research on the QSHE has primarily focused on non-magnetic materials. In this work, we extend the research scope of the QSHE to altermagnets. We establish the relevant symmetry constraints and identify all magnetic point groups that can realize the altermagnetic QSHE. Symmetry analysis reveals that pronounced spin-valley locking or spin-valley-layer locking universally exists in these systems. The concerted interaction between band inversion and spin-valley locking collectively gives rise to the helical edge states. Using first-principles calculations and theoretical models, we demonstrate that monolayer Nb2SeTeO exhibits an altermagnetic QSHE characterized by spin-valley locking, while bilayer Hf3Se3Te2 manifests an altermagnetic QSHE featuring spin-valley-layer locking. This work clarifies the intrinsic symmetry correlation between altermagnetism and quantum spin Hall topological phases, providing a brand-new theoretical perspective and research platform for exploring magnetic topological systems and developing next-generation spintronic devices

cond-mat.mes-hall↗

arXiv2026

Byzantine-Robust Distributed Sparse Learning Revisited

We revisit Byzantine robust distributed estimation for high-dimensional sparse linear models. By combining local $\ell_1$-regularized robust estimation with robust aggregation at the server, the framework applies to pseudo-Huber regression, quantile regression, and sparse SVM. We show that the resulting estimators yield non-asymptotic guarantees and attain near-optimal statistical rates under mild conditions, while remaining communication-efficient. Simulations confirm strong robustness in estimation, support recovery and classification accuracy under various Byzantine attacks.

cs.LG↗

arXiv2026

Opportunistic Bone-Loss Screening from Routine Knee Radiographs Using a Multi-Task Deep Learning Framework with Sensitivity-Constrained Threshold Optimization

Background: Osteoporosis and osteopenia are often undiagnosed until fragility fractures occur. Dual-energy X-ray absorptiometry (DXA) is the reference standard for bone mineral density (BMD) assessment, but access remains limited. Knee radiographs are obtained at high volume for osteoarthritis evaluation and may offer an opportunity for opportunistic bone-loss screening. Objective: To develop and evaluate a multi-task deep learning system for opportunistic bone-loss screening from routine knee radiographs without additional imaging or patient visits. Methods: We developed STR-Net, a multi-task framework for single-channel grayscale knee radiographs. The model includes a shared backbone, global average pooling feature aggregation, a shared neck, and a task-aware representation routing module connected to three task-specific heads: binary screening (Normal vs. Bone Loss), severity sub-classification (Osteopenia vs. Osteoporosis), and weakly coupled T-score regression with optional clinical variables. A sensitivity-constrained threshold optimization strategy (minimum sensitivity >= 0.86) was applied. The dataset included 1,570 knee radiographs, split at the patient level into training (n=1,120), validation (n=226), and test (n=224) sets. Results: On the held-out test set, STR-Net achieved an AUROC of 0.933, sensitivity of 0.904, specificity of 0.773, and AUPRC of 0.956 for binary screening. Severity sub-classification achieved an AUROC of 0.898. The T-score regression branch showed a Pearson correlation of 0.801 with DXA-measured T-scores in a pilot subset (n=31), with MAE of 0.279 and RMSE of 0.347. Conclusions: STR-Net enables single-pass bone-loss screening, severity stratification, and quantitative T-score estimation from routine knee radiographs. Prospective clinical validation is needed before deployment.

cs.CV↗

arXiv2026

Logics-Parsing-Omni Technical Report

Addressing the challenges of fragmented task definitions and the heterogeneity of unstructured data in multimodal parsing, this paper proposes the Omni Parsing framework. This framework establishes a Unified Taxonomy covering documents, images, and audio-visual streams, introducing a progressive parsing paradigm that bridges perception and cognition. Specifically, the framework integrates three hierarchical levels: 1) Holistic Detection, which achieves precise spatial-temporal grounding of objects or events to establish a geometric baseline for perception; 2) Fine-grained Recognition, which performs symbolization (e.g., OCR/ASR) and attribute extraction on localized objects to complete structured entity parsing; and 3) Multi-level Interpreting, which constructs a reasoning chain from local semantics to global logic. A pivotal advantage of this framework is its evidence anchoring mechanism, which enforces a strict alignment between high-level semantic descriptions and low-level facts. This enables ``evidence-based'' logical induction, transforming unstructured signals into standardized knowledge that is locatable, enumerable, and traceable. Building on this foundation, we constructed a standardized dataset and released the Logics-Parsing-Omni model, which successfully converts complex audio-visual signals into machine-readable structured knowledge. Experiments demonstrate that fine-grained perception and high-level cognition are synergistic, effectively enhancing model reliability. Furthermore, to quantitatively evaluate these capabilities, we introduce OmniParsingBench. Code, models and the benchmark are released at https://github.com/alibaba/Logics-Parsing/tree/master/Logics-Parsing-Omni.

cs.AI↗

arXiv2025

Eigen-1: Adaptive Multi-Agent Refinement with Monitor-Based RAG for Scientific Reasoning

Large language models (LLMs) have recently shown strong progress on scientific reasoning, yet two major bottlenecks remain. First, explicit retrieval fragments reasoning, imposing a hidden "tool tax" of extra tokens and steps. Second, multi-agent pipelines often dilute strong solutions by averaging across all candidates. We address these challenges with a unified framework that combines implicit retrieval and structured collaboration. At its foundation, a Monitor-based retrieval module operates at the token level, integrating external knowledge with minimal disruption to reasoning. On top of this substrate, Hierarchical Solution Refinement (HSR) iteratively designates each candidate as an anchor to be repaired by its peers, while Quality-Aware Iterative Reasoning (QAIR) adapts refinement to solution quality. On Humanity's Last Exam (HLE) Bio/Chem Gold, our framework achieves 48.3\% accuracy -- the highest reported to date, surpassing the strongest agent baseline by 13.4 points and leading frontier LLMs by up to 18.1 points, while simultaneously reducing token usage by 53.5\% and agent steps by 43.7\%. Results on SuperGPQA and TRQA confirm robustness across domains. Error analysis shows that reasoning failures and knowledge gaps co-occur in over 85\% of cases, while diversity analysis reveals a clear dichotomy: retrieval tasks benefit from solution variety, whereas reasoning tasks favor consensus. Together, these findings demonstrate how implicit augmentation and structured refinement overcome the inefficiencies of explicit tool use and uniform aggregation. Code is available at: https://github.com/tangxiangru/Eigen-1.

cs.CL↗

arXiv2025

Source-Free Domain Adaptive Semantic Segmentation of Remote Sensing Images with Diffusion-Guided Label Enrichment

Research on unsupervised domain adaptation (UDA) for semantic segmentation of remote sensing images has been extensively conducted. However, research on how to achieve domain adaptation in practical scenarios where source domain data is inaccessible namely, source-free domain adaptation (SFDA) remains limited. Self-training has been widely used in SFDA, which requires obtaining as many high-quality pseudo-labels as possible to train models on target domain data. Most existing methods optimize the entire pseudo-label set to obtain more supervisory information. However, as pseudo-label sets often contain substantial noise, simultaneously optimizing all labels is challenging. This limitation undermines the effectiveness of optimization approaches and thus restricts the performance of self-training. To address this, we propose a novel pseudo-label optimization framework called Diffusion-Guided Label Enrichment (DGLE), which starts from a few easily obtained high-quality pseudo-labels and propagates them to a complete set of pseudo-labels while ensuring the quality of newly generated labels. Firstly, a pseudo-label fusion method based on confidence filtering and super-resolution enhancement is proposed, which utilizes cross-validation of details and contextual information to obtain a small number of high-quality pseudo-labels as initial seeds. Then, we leverage the diffusion model to propagate incomplete seed pseudo-labels with irregular distributions due to its strong denoising capability for randomly distributed noise and powerful modeling capacity for complex distributions, thereby generating complete and high-quality pseudo-labels. This method effectively avoids the difficulty of directly optimizing the complete set of pseudo-labels, significantly improves the quality of pseudo-labels, and thus enhances the model's performance in the target domain.

cs.CV↗

arXiv2025

Investigating disc-corona interaction in axisymmetric accretion disc models

The interaction between the accretion disc and its corona plays a critical role in the energy balance and emission mechanisms in astrophysical systems such as active galactic nuclei and X-ray binaries. However, the detailed physics of disc-corona interactions, including the mechanisms driving disc evaporation and the impact of accretion rate and viscosity, remain poorly understood. Our study aims to extend the well-known disc evaporation model to investigate the disc-corona interaction in a 2D axisymmetric, time-dependent hydrodynamic model, focusing on the effects of viscosity, accretion rate, and their influence on disc evaporation, luminosity, and corona formation. We develop a hydrodynamic model consisting of a thin accretion disc, a corona, and a vacuum region. Our model is implemented in Athena++, with the gas-vacuum interface tracking algorithm to handle the vacuum regions. We perform simulations incorporating turbulent viscosity, thermal conduction, Bremsstrahlung cooling, and artificial disc cooling, starting from an adiabatic state to explore the disc-corona interaction. We demonstrate the presence of acoustic shock heating. We find that viscosity dominates the intensity of disc evaporation, that the accretion rate primarily determines the disc truncation radius and the disc luminosity, and that there may be a positive correlation between the corona luminosity and the evaporation intensity. We find the warm gas required by the warm corona model. We also compare our results with observations and simulations, and estimate the $y$-parameters to explore the potential effects of Compton cooling as well as the potential effects of the warm corona.

astro-ph.HE↗

arXiv2024

Continuity of Lyapunov exponents for stochastic differential equations

For non-autonomous linear stochastic differential equations (SDEs), we establish that the top Lyapunov exponent is continuous if the coefficients "almost" uniformly converge. For autonomous SDEs, assuming the existence of invariant measures and the convergence of coefficients and their derivatives in pointwise sense, we get the continuity of all Lyapunov exponents. Furthermore, we demonstrate that for autonomous SDEs with strict monotonicity condition, all Lyapunov exponents are Lipschitz continuous with respect to the coefficients under the $L^{p,p}$ norm ($p>2$). Similarly, the Hölder continuity of Lyapunov exponents holds under weaker regularity conditions. It seems that the continuity of Lyapunov exponents has not been studied for SDEs so far, in spite that there are many results in this direction for discrete-time dynamical systems.

math.DS↗

arXiv2024

Error bounds of Median-of-means estimators with VC-dimension

We obtain the upper error bounds of robust estimators for mean vector, using the median-of-means (MOM) method. The method is designed to handle data with heavy tails and contamination, with only a finite second moment, which is weaker than many others, relying on the VC dimension rather than the Rademacher complexity to measure statistical complexity. This allows us to implement MOM in covariance estimation, without imposing conditions such as $L$-sub-Gaussian or $L_{4}-L_{2}$ norm equivalence. In particular, we derive a new robust estimator, the MOM version of the halfspace depth, along with error bounds for mean estimation in any norm.

math.ST↗

arXiv2024

Copula-based semiparametric nonnormal transformed linear model for survival data with dependent censoring

Although the independent censoring assumption is commonly used in survival analysis, it can be violated when the censoring time is related to the survival time, which often happens in many practical applications. To address this issue, we propose a flexible semiparametric method for dependent censored data. Our approach involves fitting the survival time and the censoring time with a joint transformed linear model, where the transformed function is unspecified. This allows for a very general class of models that can account for possible covariate effects, while also accommodating administrative censoring. We assume that the transformed variables have a bivariate nonnormal distribution based on parametric copulas and parametric marginals, which further enhances the flexibility of our method. We demonstrate the identifiability of the proposed model and establish the consistency and asymptotic normality of the model parameters under appropriate regularity conditions and assumptions. Furthermore, we evaluate the performance of our method through extensive simulation studies, and provide a real data example for illustration.

stat.ME↗

arXiv2024

The multiplicative ergodic theorem for McKean-Vlasov SDEs

In this paper, we establish the multiplicative ergodic theorem for McKean-Vlasov stochastic differential equations, in which the Lyapunov exponent is defined using the upper limit. The reasonability of this definition is illustrated through an example; i.e., even when the coefficients are regular enough and their first-order derivatives are bounded, the upper limit cannot be replaced by a limit, as the limit may not exist. Furthermore, the example reveals how the dependence on distribution significantly influences the dynamics of the system and evidently distinguishes McKean-Vlasov stochastic differential equations from classical stochastic differential equations.

math.DS↗

arXiv2023

Multi-Granularity Click Confidence Learning via Self-Distillation in Recommendation

Recommendation systems rely on historical clicks to learn user interests and provide appropriate items. However, current studies tend to treat clicks equally, which may ignore the assorted intensities of user interests in different clicks. In this paper, we aim to achieve multi-granularity Click confidence Learning via Self-Distillation in recommendation (CLSD). Due to the lack of supervised signals in click confidence, we first apply self-supervised learning to obtain click confidence scores via a global self-distillation method. After that, we define a local confidence function to adapt confidence scores at the user group level, since the confidence distributions can be varied among user groups. With the combination of multi-granularity confidence learning, we can distinguish the quality of clicks and model user interests more accurately without involving extra data and model structures. The significant improvements over different backbones on industrial offline and online experiments in a real-world recommender system prove the effectiveness of our model. Recently, CLSD has been deployed on a large-scale recommender system, affecting over 400 million users.

cs.IR↗

arXiv2023

Learning from All Sides: Diversified Positive Augmentation via Self-distillation in Recommendation

Personalized recommendation relies on user historical behaviors to provide user-interested items, and thus seriously struggles with the data sparsity issue. A powerful positive item augmentation is beneficial to address the sparsity issue, while few works could jointly consider both the accuracy and diversity of these augmented training labels. In this work, we propose a novel model-agnostic Diversified self-distillation guided positive augmentation (DivSPA) for accurate and diverse positive item augmentations. Specifically, DivSPA first conducts three types of retrieval strategies to collect high-quality and diverse positive item candidates according to users' overall interests, short-term intentions, and similar users. Next, a self-distillation module is conducted to double-check and rerank these candidates as the final positive augmentations. Extensive offline and online evaluations verify the effectiveness of our proposed DivSPA on both accuracy and diversity. DivSPA is simple and effective, which could be conveniently adapted to other base models and systems. Currently, DivSPA has been deployed on multiple widely-used real-world recommender systems.

cs.IR↗

arXiv2023

CT4Rec: Simple yet Effective Consistency Training for Sequential Recommendation

Sequential recommendation methods are increasingly important in cutting-edge recommender systems. Through leveraging historical records, the systems can capture user interests and perform recommendations accordingly. State-of-the-art sequential recommendation models proposed very recently combine contrastive learning techniques for obtaining high-quality user representations. Though effective and performing well, the models based on contrastive learning require careful selection of data augmentation methods and pretext tasks, efficient negative sampling strategies, and massive hyper-parameters validation. In this paper, we propose an ultra-simple alternative for obtaining better user representations and improving sequential recommendation performance. Specifically, we present a simple yet effective \textbf{C}onsistency \textbf{T}raining method for sequential \textbf{Rec}ommendation (CT4Rec) in which only two extra training objectives are utilized without any structural modifications and data augmentation. Experiments on three benchmark datasets and one large newly crawled industrial corpus demonstrate that our proposed method outperforms SOTA models by a large margin and with much less training time than these based on contrastive learning. Online evaluation on real-world content recommendation system also achieves 2.717\% improvement on the click-through rate and 3.679\% increase on the average click number per capita. Further exploration reveals that such a simple method has great potential for CTR prediction. Our code is available at \url{https://github.com/ct4rec/CT4Rec.git}.

cs.IR↗

arXiv2023

Graph Exploration Matters: Improving both individual-level and system-level diversity in WeChat Feed Recommender

There are roughly three stages in real industrial recommendation systems, candidates generation (retrieval), ranking and reranking. Individual-level diversity and system-level diversity are both important for industrial recommender systems. The former focus on each single user's experience, while the latter focus on the difference among users. Graph-based retrieval strategies are inevitably hijacked by heavy users and popular items, leading to the convergence of candidates for users and the lack of system-level diversity. Meanwhile, in the reranking phase, Determinantal Point Process (DPP) is deployed to increase individual-level diverisity. Heavily relying on the semantic information of items, DPP suffers from clickbait and inaccurate attributes. Besides, most studies only focus on one of the two levels of diversity, and ignore the mutual influence among different stages in real recommender systems. We argue that individual-level diversity and system-level diversity should be viewed as an integrated problem, and we provide an efficient and deployable solution for web-scale recommenders. Generally, we propose to employ the retrieval graph information in diversity-based reranking, by which to weaken the hidden similarity of items exposed to users, and consequently gain more graph explorations to improve the system-level diveristy. Besides, we argue that users' propensity for diversity changes over time in content feed recommendation. Therefore, with the explored graph, we also propose to capture the user's real-time personalized propensity to the diversity. We implement and deploy the combined system in WeChat App's Top Stories used by hundreds of millions of users. Offline simulations and online A/B tests show our solution can effectively improve both user engagement and system revenue.

cs.LG↗