Shuang Zhou

Xin Wang, Liangtai Sun, Yaoming Zhu et al.

Existing code-generation benchmarks score a single mapping from a complete prompt to a one-shot output. However, real web development is different. Users seldom write a full spec at the start; many requirements only become clear once they look at an intermediate result and react to it. We present Asuka-Bench, a benchmark that pairs underspecified user intent with multi-round refinement, grounded in browser-rendered behavior. Each task is resolved through a closed loop: a Code Agent generates a web project, a UI Agent executes test cases on the deployed site, and a User LLM turns evaluation outcomes into natural-language feedback for the next round. The benchmark comprises 50 web tasks with 784 evaluation criteria and 2402 expected outcomes. We benchmark 8 LLMs across 2 agent frameworks. The results separate models clearly: weighted Task Pass Rate varies by 38 percentage points and models also differ substantially in their ability to repair from feedback. Asuka-Bench is also far from saturated: even the strongest model completes only 52% of projects after three rounds.

Meituan LongCat Team, Anchun Gui, Bei Li et al.

We introduce LongCat-Flash-Thinking-2601, a 560-billion-parameter open-source Mixture-of-Experts (MoE) reasoning model with superior agentic reasoning capability. LongCat-Flash-Thinking-2601 achieves state-of-the-art performance among open-source models on a wide range of agentic benchmarks, including agentic search, agentic tool use, and tool-integrated reasoning. Beyond benchmark performance, the model demonstrates strong generalization to complex tool interactions and robust behavior under noisy real-world environments. Its advanced capability stems from a unified training framework that combines domain-parallel expert training with subsequent fusion, together with an end-to-end co-design of data construction, environments, algorithms, and infrastructure spanning from pre-training to post-training. In particular, the model's strong generalization capability in complex tool-use are driven by our in-depth exploration of environment scaling and principled task construction. To optimize long-tailed, skewed generation and multi-turn agentic interactions, and to enable stable training across over 10,000 environments spanning more than 20 domains, we systematically extend our asynchronous reinforcement learning framework, DORA, for stable and efficient large-scale multi-environment training. Furthermore, recognizing that real-world tasks are inherently noisy, we conduct a systematic analysis and decomposition of real-world noise patterns, and design targeted training procedures to explicitly incorporate such imperfections into the training process, resulting in improved robustness for real-world applications. To further enhance performance on complex reasoning tasks, we introduce a Heavy Thinking mode that enables effective test-time scaling by jointly expanding reasoning depth and width through intensive parallel thinking.

Shuang Zhou, Zidu Xu, Mian Zhang et al.

Automatic disease diagnosis has become increasingly valuable in clinical practice. The advent of large language models (LLMs) has catalyzed a paradigm shift in artificial intelligence, with growing evidence supporting the efficacy of LLMs in diagnostic tasks. Despite the increasing attention in this field, a holistic view is still lacking. Many critical aspects remain unclear, such as the diseases and clinical data to which LLMs have been applied, the LLM techniques employed, and the evaluation methods used. In this article, we perform a comprehensive review of LLM-based methods for disease diagnosis. Our review examines the existing literature across various dimensions, including disease types and associated clinical specialties, clinical data, LLM techniques, and evaluation methods. Additionally, we offer recommendations for applying and evaluating LLMs for diagnostic tasks. Furthermore, we assess the limitations of current research and discuss future directions. To our knowledge, this is the first comprehensive review for LLM-based disease diagnosis.

Shuang Zhou, Xiao Huang, Ninghao Liu et al.

Graph anomaly detection (GAD) is a vital task since even a few anomalies can pose huge threats to benign users. Recent semi-supervised GAD methods, which can effectively leverage the available labels as prior knowledge, have achieved superior performances than unsupervised methods. In practice, people usually need to identify anomalies on new (sub)graphs to secure their business, but they may lack labels to train an effective detection model. One natural idea is to directly adopt a trained GAD model to the new (sub)graph for testing. However, we find that existing semi-supervised GAD methods suffer from poor generalization issue, i.e., well-trained models could not perform well on an unseen area (i.e., not accessible in training) of the same graph. It may cause great troubles. In this paper, we base on the phenomenon and propose a general and novel research problem of generalized graph anomaly detection that aims to effectively identify anomalies on both the training-domain graph and unseen testing graph to eliminate potential dangers. Nevertheless, it is a challenging task since only limited labels are available, and the normal background may differ between training and testing data. Accordingly, we propose a data augmentation method named \textit{AugAN} (\uline{Aug}mentation for \uline{A}nomaly and \uline{N}ormal distributions) to enrich training data and boost the generalizability of GAD models. Experiments verify the effectiveness of our method in improving model generalizability.

Shuang Zhou, Xiao Huang, Ninghao Liu et al.

Graph anomaly detection (GAD) is a vital task since even a few anomalies can pose huge threats to benign users. Recent semi-supervised GAD methods, which can effectively leverage the available labels as prior knowledge, have achieved superior performances than unsupervised methods. In practice, people usually need to identify anomalies on new (sub)graphs to secure their business, but they may lack labels to train an effective detection model. One natural idea is to directly adopt a trained GAD model to the new (sub)graph for testing. However, we find that existing semi-supervised GAD methods suffer from poor generalization issue, i.e., well-trained models could not perform well on an unseen area (i.e., not accessible in training) of the same graph. It may cause great troubles. In this paper, we base on the phenomenon and propose a general and novel research problem of generalized graph anomaly detection that aims to effectively identify anomalies on both the training-domain graph and unseen testing graph to eliminate potential dangers. Nevertheless, it is a challenging task since only limited labels are available, and the normal background may differ between training and testing data. Accordingly, we propose a data augmentation method named \textit{AugAN} (\uline{Aug}mentation for \uline{A}nomaly and \uline{N}ormal distributions) to enrich training data and boost the generalizability of GAD models. Experiments verify the effectiveness of our method in improving model generalizability.

DeepSeek-AI, Daya Guo, Dejian Yang et al. · stanford, tsinghua

We introduce our first-generation reasoning models, DeepSeek-R1-Zero and DeepSeek-R1. DeepSeek-R1-Zero, a model trained via large-scale reinforcement learning (RL) without supervised fine-tuning (SFT) as a preliminary step, demonstrates remarkable reasoning capabilities. Through RL, DeepSeek-R1-Zero naturally emerges with numerous powerful and intriguing reasoning behaviors. However, it encounters challenges such as poor readability, and language mixing. To address these issues and further enhance reasoning performance, we introduce DeepSeek-R1, which incorporates multi-stage training and cold-start data before RL. DeepSeek-R1 achieves performance comparable to OpenAI-o1-1217 on reasoning tasks. To support the research community, we open-source DeepSeek-R1-Zero, DeepSeek-R1, and six dense models (1.5B, 7B, 8B, 14B, 32B, 70B) distilled from DeepSeek-R1 based on Qwen and Llama.

DeepSeek-AI, Aixin Liu, Bei Feng et al. · pku

We present DeepSeek-V2, a strong Mixture-of-Experts (MoE) language model characterized by economical training and efficient inference. It comprises 236B total parameters, of which 21B are activated for each token, and supports a context length of 128K tokens. DeepSeek-V2 adopts innovative architectures including Multi-head Latent Attention (MLA) and DeepSeekMoE. MLA guarantees efficient inference through significantly compressing the Key-Value (KV) cache into a latent vector, while DeepSeekMoE enables training strong models at an economical cost through sparse computation. Compared with DeepSeek 67B, DeepSeek-V2 achieves significantly stronger performance, and meanwhile saves 42.5% of training costs, reduces the KV cache by 93.3%, and boosts the maximum generation throughput to 5.76 times. We pretrain DeepSeek-V2 on a high-quality and multi-source corpus consisting of 8.1T tokens, and further perform Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL) to fully unlock its potential. Evaluation results show that, even with only 21B activated parameters, DeepSeek-V2 and its chat versions still achieve top-tier performance among open-source models.

Zaifu Zhan, Min Zeng, Shuang Zhou et al.

Objective: To improve the efficiency of medical question answering (MedQA) with large language models (LLMs) by avoiding unnecessary reasoning while maintaining accuracy. Methods: We propose Selective Chain-of-Thought (Selective CoT), an inference-time strategy that first predicts whether a question requires reasoning and generates a rationale only when needed. Two open-source LLMs (Llama-3.1-8B and Qwen-2.5-7B) were evaluated on four biomedical QA benchmarks-HeadQA, MedQA-USMLE, MedMCQA, and PubMedQA. Metrics included accuracy, total generated tokens, and inference time. Results: Selective CoT reduced inference time by 13-45% and token usage by 8-47% with minimal accuracy loss ($\leq$4\%). In some model-task pairs, it achieved both higher accuracy and greater efficiency than standard CoT. Compared with fixed-length CoT, Selective CoT reached similar or superior accuracy at substantially lower computational cost. Discussion: Selective CoT dynamically balances reasoning depth and efficiency by invoking explicit reasoning only when beneficial, reducing redundancy on recall-type questions while preserving interpretability. Conclusion: Selective CoT provides a simple, model-agnostic, and cost-effective approach for medical QA, aligning reasoning effort with question complexity to enhance real-world deployability of LLM-based clinical systems.

Pan Xiao, Peijie Qiu, Sungmin Ha et al.

Learning rich data representations from unlabeled data is a key challenge towards applying deep learning algorithms in downstream tasks. Several variants of variational autoencoders (VAEs) have been proposed to learn compact data representations by encoding high-dimensional data in a lower dimensional space. Two main classes of VAEs methods may be distinguished depending on the characteristics of the meta-priors that are enforced in the representation learning step. The first class of methods derives a continuous encoding by assuming a static prior distribution in the latent space. The second class of methods learns instead a discrete latent representation using vector quantization (VQ) along with a codebook. However, both classes of methods suffer from certain challenges, which may lead to suboptimal image reconstruction results. The first class suffers from posterior collapse, whereas the second class suffers from codebook collapse. To address these challenges, we introduce a new VAE variant, termed sparse coding-based VAE with learned ISTA (SC-VAE), which integrates sparse coding within variational autoencoder framework. The proposed method learns sparse data representations that consist of a linear combination of a small number of predetermined orthogonal atoms. The sparse coding problem is solved using a learnable version of the iterative shrinkage thresholding algorithm (ISTA). Experiments on two image datasets demonstrate that our model achieves improved image reconstruction results compared to state-of-the-art methods. Moreover, we demonstrate that the use of learned sparse code vectors allows us to perform downstream tasks like image generation and unsupervised image segmentation through clustering image patches.

Shuang Zhou, Kai Yu, Song Wang et al.

Heart diseases remain a leading cause of morbidity and mortality worldwide, necessitating accurate and trustworthy differential diagnosis. However, existing artificial intelligence-based diagnostic methods are often limited by insufficient cardiology knowledge, inadequate support for complex reasoning, and poor interpretability. Here we present HeartAgent, a cardiology-specific agent system designed to support a reliable and explainable differential diagnosis. HeartAgent integrates customized tools and curated data resources and orchestrates multiple specialized sub-agents to perform complex reasoning while generating transparent reasoning trajectories and verifiable supporting references. Evaluated on the MIMIC dataset and a private electronic health records cohort, HeartAgent achieved over 36% and 20% improvements over established comparative methods, in top-3 diagnostic accuracy, respectively. Additionally, clinicians assisted by HeartAgent demonstrated gains of 26.9% in diagnostic accuracy and 22.7% in explanatory quality compared with unaided experts. These results demonstrate that HeartAgent provides reliable, explainable, and clinically actionable decision support for cardiovascular care.

DeepSeek-AI, Aixin Liu, Aoxue Mei et al.

We introduce DeepSeek-V3.2, a model that harmonizes high computational efficiency with superior reasoning and agent performance. The key technical breakthroughs of DeepSeek-V3.2 are as follows: (1) DeepSeek Sparse Attention (DSA): We introduce DSA, an efficient attention mechanism that substantially reduces computational complexity while preserving model performance in long-context scenarios. (2) Scalable Reinforcement Learning Framework: By implementing a robust reinforcement learning protocol and scaling post-training compute, DeepSeek-V3.2 performs comparably to GPT-5. Notably, our high-compute variant, DeepSeek-V3.2-Speciale, surpasses GPT-5 and exhibits reasoning proficiency on par with Gemini-3.0-Pro, achieving gold-medal performance in both the 2025 International Mathematical Olympiad (IMO) and the International Olympiad in Informatics (IOI). (3) Large-Scale Agentic Task Synthesis Pipeline: To integrate reasoning into tool-use scenarios, we developed a novel synthesis pipeline that systematically generates training data at scale. This methodology facilitates scalable agentic post-training, yielding substantial improvements in generalization and instruction-following robustness within complex, interactive environments.

Zhu Wang, Junnan Dong, Shuang Zhou et al.

Graph anomaly detection (GAD) is widely applied in many areas, such as financial fraud detection and social spammer detection. Anomalous nodes in the graph not only impact their own communities but also create a ripple effect on neighbors throughout the graph structure. Detecting anomalous nodes in complex graphs has been a challenging task. While existing GAD methods assume all labels are correct, real-world scenarios often involve inaccurate annotations. These noisy labels can severely degrade GAD performance because, with anomalies representing a minority class, even a small number of mislabeled instances can disproportionately interfere with detection models. Cutting edges to mitigate the negative effects of noisy labels is a good option; however, it has both positive and negative influences and also presents an issue of weak supervision. To perform effective GAD with noisy labels, we propose REinforced Graph Anomaly Detector (REGAD) by pruning the edges of candidate nodes potentially with mistaken labels. Moreover, we design the performance feedback based on strategically crafted confident labels to guide the cutting process, ensuring optimal results. Specifically, REGAD contains two novel components. (i) A tailored policy network, which involves two-step actions to remove negative effect propagation step by step. (ii) A policy-in-the-loop mechanism to identify suitable edge removal strategies that control the propagation of noise on the graph and estimate the updated structure to obtain reliable pseudo labels iteratively. Experiments on three real-world datasets demonstrate that REGAD outperforms all baselines under different noisy ratios.

Junlin Liu, Shengnan An, Shuang Zhou et al.

Contemporary large language models (LLMs) have demonstrated remarkable reasoning capabilities, particularly in specialized domains like mathematics and physics. However, their ability to generalize these reasoning skills to more general and broader contexts--often termed general reasoning--remains under-explored. Unlike domain-specific reasoning, general reasoning relies less on expert knowledge but still presents formidable reasoning challenges, such as complex constraints, nested logical branches, and semantic interference. To address this gap, we introduce General365, a benchmark specifically designed to assess general reasoning in LLMs. By restricting background knowledge to a K-12 level, General365 explicitly decouples reasoning from specialized expertise. The benchmark comprises 365 seed problems and 1,095 variant problems across eight categories, ensuring both high difficulty and diversity. Evaluations across 26 leading LLMs reveal that even the top-performing model achieves only 62.8% accuracy, in stark contrast to the near-perfect performances of LLMs in math and physics benchmarks. These results suggest that the reasoning abilities of current LLMs are heavily domain-dependent, leaving significant room for improvement in broader applications. We envision General365 as a catalyst for advancing LLM reasoning beyond domain-specific tasks toward robust, general-purpose real-world scenarios. Code, Dataset, and Leaderboard: https://general365.github.io

Shuang Zhou, Kai Yu, Zaifu Zhan et al.

Epilepsy and psychogenic non-epileptic seizures often present with similar seizure-like manifestations but require fundamentally different management strategies. Misdiagnosis is common and can lead to prolonged diagnostic delays, unnecessary treatments, and substantial patient morbidity. Although prolonged video-electroencephalography is the diagnostic gold standard, its high cost and limited accessibility hinder timely diagnosis. Here, we developed a low-cost, effective approach, EpiScreen, for early epilepsy detection by utilizing routinely collected clinical notes from electronic health records. Through fine-tuning large language models on labeled notes, EpiScreen achieved an AUC of up to 0.875 on the MIMIC-IV dataset and 0.980 on a private cohort of the University of Minnesota. In a clinician-AI collaboration setting, EpiScreen-assisted neurologists outperformed unaided experts by up to 10.9%. Overall, this study demonstrates that EpiScreen supports early epilepsy detection, facilitating timely and cost-effective screening that may reduce diagnostic delays and avoid unnecessary interventions, particularly in resource-limited regions.

Shengnan An, Xunliang Cai, Xuezhi Cao et al.

We present AMO-Bench, an Advanced Mathematical reasoning benchmark with Olympiad level or even higher difficulty, comprising 50 human-crafted problems. Existing benchmarks have widely leveraged high school math competitions for evaluating mathematical reasoning capabilities of large language models (LLMs). However, many existing math competitions are becoming less effective for assessing top-tier LLMs due to performance saturation (e.g., AIME24/25). To address this, AMO-Bench introduces more rigorous challenges by ensuring all 50 problems are (1) cross-validated by experts to meet at least the International Mathematical Olympiad (IMO) difficulty standards, and (2) entirely original problems to prevent potential performance leakages from data memorization. Moreover, each problem in AMO-Bench requires only a final answer rather than a proof, enabling automatic and robust grading for evaluation. Experimental results across 26 LLMs on AMO-Bench show that even the best-performing model achieves only 52.4% accuracy on AMO-Bench, with most LLMs scoring below 40%. Beyond these poor performances, our further analysis reveals a promising scaling trend with increasing test-time compute on AMO-Bench. These results highlight the significant room for improving the mathematical reasoning in current LLMs. We release AMO-Bench to facilitate further research into advancing the reasoning abilities of language models. https://amo-bench.github.io/

Shuang Zhou, Mingquan Lin, Sirui Ding et al.

Automatic differential diagnosis (DDx) is an essential medical task that generates a list of potential diseases as differentials based on patient symptom descriptions. In practice, interpreting these differential diagnoses yields significant value but remains under-explored. Given the powerful capabilities of large language models (LLMs), we investigated using LLMs for interpretable DDx. Specifically, we curated the first DDx dataset with expert-derived interpretation on 570 clinical notes. Besides, we proposed Dual-Inf, a novel framework that enabled LLMs to conduct bidirectional inference (i.e., from symptoms to diagnoses and vice versa) for DDx interpretation. Both human and automated evaluation validated its efficacy in predicting and elucidating differentials across four base LLMs. In addition, Dual-Inf could reduce interpretation errors and hold promise for rare disease explanations. To the best of our knowledge, it is the first work that customizes LLMs for DDx explanation and comprehensively evaluates their interpretation performance. Overall, our study bridges a critical gap in DDx interpretation and enhances clinical decision-making.

Kai Yu, Shuang Zhou, Yiran Song et al.

Multimodal self-supervised pretraining offers a promising route to cancer prognosis by integrating histopathology whole-slide images, gene expression, and pathology reports, yet most existing approaches require fully paired and complete inputs. In practice, clinical cohorts are fragmented and often miss one or more modalities, limiting both supervised fusion and scalable multimodal pretraining. We propose PRIME, a missing-aware multimodal self-supervised pretraining framework that learns robust and transferable representations from partially observed cohorts. PRIME maps heterogeneous modality embeddings into a unified token space and introduces a shared prototype memory bank for latent-space semantic imputation via patient-level consensus retrieval, producing structurally aligned tokens without reconstructing raw signals. Two complementary pretraining objectives: inter-modality alignment and post-fusion consistency under structured missingness augmentation, jointly learn representations that remain predictive under arbitrary modality subsets. We evaluate PRIME on The Cancer Genome Atlas with label-free pretraining on 32 cancer types and downstream 5-fold evaluation on five cohorts across overall survival prediction, 3-year mortality classification, and 3-year recurrence classification. PRIME achieves the best macro-average performance among all compared methods, reaching 0.653 C-index, 0.689 AUROC, and 0.637 AUROC on the three tasks, respectively, while improving robustness under test-time missingness and supporting parameter-efficient and label-efficient adaptation. These results support missing-aware multimodal pretraining as a practical strategy for prognosis modeling in fragmented clinical data settings.

Shuang Zhou, Jiashuo Wang, Zidu Xu et al.

Explainable disease diagnosis, which leverages patient information (e.g., signs and symptoms) and computational models to generate probable diagnoses and reasonings, offers clear clinical values. However, when clinical notes encompass insufficient evidence for a definite diagnosis, such as the absence of definitive symptoms, diagnostic uncertainty usually arises, increasing the risk of misdiagnosis and adverse outcomes. Although explicitly identifying and explaining diagnostic uncertainties is essential for trustworthy diagnostic systems, it remains under-explored. To fill this gap, we introduce ConfiDx, an uncertainty-aware large language model (LLM) created by fine-tuning open-source LLMs with diagnostic criteria. We formalized the task and assembled richly annotated datasets that capture varying degrees of diagnostic ambiguity. Evaluating ConfiDx on real-world datasets demonstrated that it excelled in identifying diagnostic uncertainties, achieving superior diagnostic performance, and generating trustworthy explanations for diagnoses and uncertainties. To our knowledge, this is the first study to jointly address diagnostic uncertainty recognition and explanation, substantially enhancing the reliability of automatic diagnostic systems.

Min Zeng, Shuang Zhou, Zaifu Zhan et al.

Medical language models must be updated as evidence and terminology evolve, yet sequential updating can trigger catastrophic forgetting. Although biomedical NLP has many static benchmarks, no unified, task-diverse benchmark exists for evaluating continual learning under standardized protocols, robustness to task order and compute-aware reporting. We introduce MedCL-Bench, which streams ten biomedical NLP datasets spanning five task families and evaluates eleven continual learning strategies across eight task orders, reporting retention, transfer, and GPU-hour cost. Across backbones and task orders, direct sequential fine-tuning on incoming tasks induces catastrophic forgetting, causing update-induced performance regressions on prior tasks. Continual learning methods occupy distinct retention-compute frontiers: parameter-isolation provides the best retention per GPU-hour, replay offers strong protection at higher cost, and regularization yields limited benefit. Forgetting is task-dependent, with multi-label topic classification most vulnerable and constrained-output tasks more robust. MedCL-Bench provides a reproducible framework for auditing model updates before deployment.

Meituan LongCat Team, Anchun Gui, Bei Li et al.

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.

Yiran Song, Yikai Zhang, Shuang Zhou et al.

Multiple instance learning (MIL) has emerged as the dominant paradigm for whole slide image (WSI) analysis in computational pathology, achieving strong diagnostic performance through patch-level feature aggregation. However, existing MIL methods face critical limitations: (1) they rely on attention mechanisms that lack causal interpretability, and (2) they fail to integrate patient demographics (age, gender, race), leading to fairness concerns across diverse populations. These shortcomings hinder clinical translation, where algorithmic bias can exacerbate health disparities. We introduce \textbf{MeCaMIL}, a causality-aware MIL framework that explicitly models demographic confounders through structured causal graphs. Unlike prior approaches treating demographics as auxiliary features, MeCaMIL employs principled causal inference -- leveraging do-calculus and collider structures -- to disentangle disease-relevant signals from spurious demographic correlations. Extensive evaluation on three benchmarks demonstrates state-of-the-art performance across CAMELYON16 (ACC/AUC/F1: 0.939/0.983/0.946), TCGA-Lung (0.935/0.979/0.931), and TCGA-Multi (0.977/0.993/0.970, five cancer types). Critically, MeCaMIL achieves superior fairness -- demographic disparity variance drops by over 65% relative reduction on average across attributes, with notable improvements for underserved populations. The framework generalizes to survival prediction (mean C-index: 0.653, +0.017 over best baseline across five cancer types). Ablation studies confirm causal graph structure is essential -- alternative designs yield 0.048 lower accuracy and 4.2x times worse fairness. These results establish MeCaMIL as a principled framework for fair, interpretable, and clinically actionable AI in digital pathology. Code will be released upon acceptance.

Shuang Zhou, Daochen Zha, Xiao Shen et al.

Time series self-supervised learning (SSL) aims to exploit unlabeled data for pre-training to mitigate the reliance on labels. Despite the great success in recent years, there is limited discussion on the potential noise in the time series, which can severely impair the performance of existing SSL methods. To mitigate the noise, the de facto strategy is to apply conventional denoising methods before model training. However, this pre-processing approach may not fully eliminate the effect of noise in SSL for two reasons: (i) the diverse types of noise in time series make it difficult to automatically determine suitable denoising methods; (ii) noise can be amplified after mapping raw data into latent space. In this paper, we propose denoising-aware contrastive learning (DECL), which uses contrastive learning objectives to mitigate the noise in the representation and automatically selects suitable denoising methods for every sample. Extensive experiments on various datasets verify the effectiveness of our method. The code is open-sourced.

DeepSeek-AI, Aixin Liu, Bei Feng et al. · stanford, tsinghua

We present DeepSeek-V3, a strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token. To achieve efficient inference and cost-effective training, DeepSeek-V3 adopts Multi-head Latent Attention (MLA) and DeepSeekMoE architectures, which were thoroughly validated in DeepSeek-V2. Furthermore, DeepSeek-V3 pioneers an auxiliary-loss-free strategy for load balancing and sets a multi-token prediction training objective for stronger performance. We pre-train DeepSeek-V3 on 14.8 trillion diverse and high-quality tokens, followed by Supervised Fine-Tuning and Reinforcement Learning stages to fully harness its capabilities. Comprehensive evaluations reveal that DeepSeek-V3 outperforms other open-source models and achieves performance comparable to leading closed-source models. Despite its excellent performance, DeepSeek-V3 requires only 2.788M H800 GPU hours for its full training. In addition, its training process is remarkably stable. Throughout the entire training process, we did not experience any irrecoverable loss spikes or perform any rollbacks. The model checkpoints are available at https://github.com/deepseek-ai/DeepSeek-V3.

Rongyang Wang, Shuang Zhou, Jiashuo Wang et al.

Multimodal large language models (MLLMs) have emerged as a promising paradigm for dental image analysis. However, their ability to capture the multi-level cognitive processes required for radiographic analysis remains unclear. Here, we present a comprehensive benchmark to evaluate the cognitive capabilities of MLLMs in dental radiographic analysis. It spans three critical imaging modalities, i.e., periapical, panoramic, and lateral cephalometric radiographs, and defines four cognitive categories: perception, comprehension, prediction, and decision-making. The benchmark comprises 27 clinically grounded tasks derived from public datasets, with manually curated annotations and 3,820 clinician assessments for evaluation. Six frontier MLLMs, including GPT-5.2 and GLM-4.6, are evaluated. We demonstrate the performance gap between MLLMs and clinicians in dental practice, delineate model strengths and limitations, characterize failure patterns, and provide recommendations for improvement. This data resource will facilitate the development of next-generation artificial intelligence systems aligned with clinical cognition, safety requirements, and workflow complexity in dental practice.

Zaifu Zhan, Shuang Zhou, Mingchen Li et al.

\textbf{Objective:} We aimed to develop an advanced multi-task large language model (LLM) framework to extract multiple types of information about dietary supplements (DS) from clinical records. \textbf{Methods:} We used four core DS information extraction tasks - namely, named entity recognition (NER: 2,949 clinical sentences), relation extraction (RE: 4,892 sentences), triple extraction (TE: 2,949 sentences), and usage classification (UC: 2,460 sentences) as our multitasks. We introduced a novel Retrieval-Augmented Multi-task Information Extraction (RAMIE) Framework, including: 1) employed instruction fine-tuning techniques with task-specific prompts, 2) trained LLMs for multiple tasks with improved storage efficiency and lower training costs, and 3) incorporated retrieval augmentation generation (RAG) techniques by retrieving similar examples from the training set. We compared RAMIE's performance to LLMs with instruction fine-tuning alone and conducted an ablation study to assess the contributions of multi-task learning and RAG to improved multitasking performance. \textbf{Results:} With the aid of the RAMIE framework, Llama2-13B achieved an F1 score of 87.39 (3.51\% improvement) on the NER task and demonstrated outstanding performance on the RE task with an F1 score of 93.74 (1.15\% improvement). For the TE task, Llama2-7B scored 79.45 (14.26\% improvement), and MedAlpaca-7B achieved the highest F1 score of 93.45 (0.94\% improvement) on the UC task. The ablation study revealed that while MTL increased efficiency with a slight trade-off in performance, RAG significantly boosted overall accuracy. \textbf{Conclusion:} This study presents a novel RAMIE framework that demonstrates substantial improvements in multi-task information extraction for DS-related data from clinical records. Our framework can potentially be applied to other domains.

Zaifu Zhan, Jun Wang, Shuang Zhou et al.

Objective: To optimize in-context learning in biomedical natural language processing by improving example selection. Methods: We introduce a novel multi-mode retrieval-augmented generation (MMRAG) framework, which integrates four retrieval strategies: (1) Random Mode, selecting examples arbitrarily; (2) Top Mode, retrieving the most relevant examples based on similarity; (3) Diversity Mode, ensuring variation in selected examples; and (4) Class Mode, selecting category-representative examples. This study evaluates MMRAG on three core biomedical NLP tasks: Named Entity Recognition (NER), Relation Extraction (RE), and Text Classification (TC). The datasets used include BC2GM for gene and protein mention recognition (NER), DDI for drug-drug interaction extraction (RE), GIT for general biomedical information extraction (RE), and HealthAdvice for health-related text classification (TC). The framework is tested with two large language models (Llama2-7B, Llama3-8B) and three retrievers (Contriever, MedCPT, BGE-Large) to assess performance across different retrieval strategies. Results: The results from the Random mode indicate that providing more examples in the prompt improves the model's generation performance. Meanwhile, Top mode and Diversity mode significantly outperform Random mode on the RE (DDI) task, achieving an F1 score of 0.9669, a 26.4% improvement. Among the three retrievers tested, Contriever outperformed the other two in a greater number of experiments. Additionally, Llama 2 and Llama 3 demonstrated varying capabilities across different tasks, with Llama 3 showing a clear advantage in handling NER tasks. Conclusion: MMRAG effectively enhances biomedical in-context learning by refining example selection, mitigating data scarcity issues, and demonstrating superior adaptability for NLP-driven healthcare applications.

Song Wang, Zhen Tan, Zihan Chen et al.

Recent progress in large language model (LLM)-based multi-agent collaboration highlights the power of structured communication in enabling collective intelligence. However, existing methods largely rely on static or graph-based inter-agent topologies, lacking the potential adaptability and flexibility in communication. In this work, we propose a new framework that rethinks multi-agent coordination through a sequential structure rather than a graph structure, offering a significantly larger topology space for multi-agent communication. Our method focuses on two key directions: (1) Next-Agent Prediction, which selects the most suitable agent role at each step, and (2) Next-Context Selection (NCS), which enables each agent to selectively access relevant information from any previous step. Together, these components construct task-adaptive communication pipelines that support both role flexibility and global information flow. Extensive evaluations across multiple benchmarks demonstrate that our approach achieves superior performance while substantially reducing communication overhead.

Zaifu Zhan, Shuang Zhou, Xiaoshan Zhou et al.

Objectives: We aim to dynamically retrieve informative demonstrations, enhancing in-context learning in multimodal large language models (MLLMs) for disease classification. Methods: We propose a Retrieval-Augmented In-Context Learning (RAICL) framework, which integrates retrieval-augmented generation (RAG) and in-context learning (ICL) to adaptively select demonstrations with similar disease patterns, enabling more effective ICL in MLLMs. Specifically, RAICL examines embeddings from diverse encoders, including ResNet, BERT, BioBERT, and ClinicalBERT, to retrieve appropriate demonstrations, and constructs conversational prompts optimized for ICL. We evaluated the framework on two real-world multi-modal datasets (TCGA and IU Chest X-ray), assessing its performance across multiple MLLMs (Qwen, Llava, Gemma), embedding strategies, similarity metrics, and varying numbers of demonstrations. Results: RAICL consistently improved classification performance. Accuracy increased from 0.7854 to 0.8368 on TCGA and from 0.7924 to 0.8658 on IU Chest X-ray. Multi-modal inputs outperformed single-modal ones, with text-only inputs being stronger than images alone. The richness of information embedded in each modality will determine which embedding model can be used to get better results. Few-shot experiments showed that increasing the number of retrieved examples further enhanced performance. Across different similarity metrics, Euclidean distance achieved the highest accuracy while cosine similarity yielded better macro-F1 scores. RAICL demonstrated consistent improvements across various MLLMs, confirming its robustness and versatility. Conclusions: RAICL provides an efficient and scalable approach to enhance in-context learning in MLLMs for multimodal disease classification.

Zaifu Zhan, Shuang Zhou, Huixue Zhou et al.

The advancement of Large Language Models (LLMs) has significantly impacted biomedical Natural Language Processing (NLP), enhancing tasks such as named entity recognition, relation extraction, event extraction, and text classification. In this context, the DeepSeek series of models have shown promising potential in general NLP tasks, yet their capabilities in the biomedical domain remain underexplored. This study evaluates multiple DeepSeek models (Distilled-DeepSeek-R1 series and Deepseek-LLMs) across four key biomedical NLP tasks using 12 datasets, benchmarking them against state-of-the-art alternatives (Llama3-8B, Qwen2.5-7B, Mistral-7B, Phi-4-14B, Gemma-2-9B). Our results reveal that while DeepSeek models perform competitively in named entity recognition and text classification, challenges persist in event and relation extraction due to precision-recall trade-offs. We provide task-specific model recommendations and highlight future research directions. This evaluation underscores the strengths and limitations of DeepSeek models in biomedical NLP, guiding their future deployment and optimization.

Yu Hou, Zaifu Zhan, Min Zeng et al.

Biomedical literature and clinical narratives pose multifaceted challenges for natural language understanding, from precise entity extraction and document synthesis to multi-step diagnostic reasoning. This study extends a unified benchmark to evaluate GPT-5 and GPT-4o under zero-, one-, and five-shot prompting across five core biomedical NLP tasks: named entity recognition, relation extraction, multi-label document classification, summarization, and simplification, and nine expanded biomedical QA datasets covering factual knowledge, clinical reasoning, and multimodal visual understanding. Using standardized prompts, fixed decoding parameters, and consistent inference pipelines, we assessed model performance, latency, and token-normalized cost under official pricing. GPT-5 consistently outperformed GPT-4o, with the largest gains on reasoning-intensive datasets such as MedXpertQA and DiagnosisArena and stable improvements in multimodal QA. In core tasks, GPT-5 achieved better chemical NER and ChemProt scores but remained below domain-tuned baselines for disease NER and summarization. Despite producing longer outputs, GPT-5 showed comparable latency and 30 to 50 percent lower effective cost per correct prediction. Fine-grained analyses revealed improvements in diagnosis, treatment, and reasoning subtypes, whereas boundary-sensitive extraction and evidence-dense summarization remain challenging. Overall, GPT-5 approaches deployment-ready performance for biomedical QA while offering a favorable balance of accuracy, interpretability, and economic efficiency. The results support a tiered prompting strategy: direct prompting for large-scale or cost-sensitive applications, and chain-of-thought scaffolds for analytically complex or high-stakes scenarios, highlighting the continued need for hybrid solutions where precision and factual fidelity are critical.

Shuang Zhou, Wenya Xie, Jiaxi Li et al.

As large language models (LLMs) become increasingly integrated into clinical decision-making, ensuring transparent and trustworthy reasoning is essential. However, existing evaluation strategies of LLMs' medical reasoning capability either suffer from unsatisfactory assessment or poor scalability, and a rigorous benchmark remains lacking. To address this, we introduce MedThink-Bench, a benchmark designed for rigorous, explainable, and scalable assessment of LLMs' medical reasoning. MedThink-Bench comprises 500 challenging questions across ten medical domains, each annotated with expert-crafted step-by-step rationales. Building on this, we propose LLM-w-Ref, a novel evaluation framework that leverages fine-grained rationales and LLM-as-a-Judge mechanisms to assess intermediate reasoning with expert-level fidelity while maintaining scalability. Experiments show that LLM-w-Ref exhibits a strong positive correlation with expert judgments. Benchmarking twelve state-of-the-art LLMs, we find that smaller models (e.g., MedGemma-27B) can surpass larger proprietary counterparts (e.g., OpenAI-o3). Overall, MedThink-Bench offers a foundational tool for evaluating LLMs' medical reasoning, advancing their safe and responsible deployment in clinical practice.

Zaifu Zhan, Shuang Zhou, Huixue Zhou et al.

Foundation models, including language models, e.g., GPT, and vision models, e.g., CLIP, have significantly advanced numerous biomedical tasks. Despite these advancements, the high inference latency and the "overthinking" issues in model inference impair the efficiency and effectiveness of foundation models, thus limiting their application in real-time clinical settings. To address these challenges, we proposed EPEE (Entropy- and Patience-based Early Exiting), a novel hybrid strategy designed to improve the inference efficiency of foundation models. The core idea was to leverage the strengths of entropy-based and patience-based early exiting methods to overcome their respective weaknesses. To evaluate EPEE, we conducted experiments on three core biomedical tasks-classification, relation extraction, and event extraction-using four foundation models (BERT, ALBERT, GPT-2, and ViT) across twelve datasets, including clinical notes and medical images. The results showed that EPEE significantly reduced inference time while maintaining or improving accuracy, demonstrating its adaptability to diverse datasets and tasks. EPEE addressed critical barriers to deploying foundation models in healthcare by balancing efficiency and effectiveness. It potentially provided a practical solution for real-time clinical decision-making with foundation models, supporting reliable and efficient workflows.

Jiong Wu, Shuang Zhou, Li Lin et al.

Diffeomorphic image registration is a fundamental step in medical image analysis, owing to its capability to ensure the invertibility of transformations and preservation of topology. Currently, unsupervised learning-based registration techniques primarily extract features at the image level, potentially limiting their efficacy. This paper proposes a novel unsupervised learning-based fully convolutional network (FCN) framework for fast diffeomorphic image registration, emphasizing feature acquisition at the image patch level. Furthermore, a novel differential operator is introduced and integrated into the FCN architecture for parameter learning. Experiments are conducted on three distinct T1-weighted magnetic resonance imaging (T1w MRI) datasets. Comparative analyses with three state-of-the-art diffeomorphic image registration approaches including a typical conventional registration algorithm and two representative unsupervised learning-based methods, reveal that the proposed method exhibits superior performance in both registration accuracy and topology preservation.

Zaifu Zhan, Shuang Zhou, Min Zeng et al.

Large language models have demonstrated remarkable capabilities in biomedical natural language processing, yet their rapid growth in size and computational requirements present a major barrier to adoption in healthcare settings where data privacy precludes cloud deployment and resources are limited. In this study, we systematically evaluated the impact of quantization on 12 state-of-the-art large language models, including both general-purpose and biomedical-specific models, across eight benchmark datasets covering four key tasks: named entity recognition, relation extraction, multi-label classification, and question answering. We show that quantization substantially reduces GPU memory requirements-by up to 75%-while preserving model performance across diverse tasks, enabling the deployment of 70B-parameter models on 40GB consumer-grade GPUs. In addition, domain-specific knowledge and responsiveness to advanced prompting methods are largely maintained. These findings provide significant practical and guiding value, highlighting quantization as a practical and effective strategy for enabling the secure, local deployment of large yet high-capacity language models in biomedical contexts, bridging the gap between technical advances in AI and real-world clinical translation.

Jie Peng, Shuang Zhou, Longwei Yang et al.

Cancer prognosis is a critical task that involves predicting patient outcomes and survival rates. To enhance prediction accuracy, previous studies have integrated diverse data modalities, such as clinical notes, medical images, and genomic data, leveraging their complementary information. However, existing approaches face two major limitations. First, they struggle to incorporate newly arrived data with varying distributions into training, such as patient records from different hospitals, thus rendering sub-optimal generalizability and limited utility in real-world applications. Second, most multimodal integration methods rely on simplistic concatenation or task-specific pipelines, which fail to capture the complex interdependencies across modalities. To address these, we propose a continually evolving multi-modal foundation model. Extensive experiments on the TCGA dataset demonstrate the effectiveness of our approach, highlighting its potential to advance cancer prognosis by enabling robust and adaptive multimodal integration.

Zhi Chang, Chukwudi Obite, Shuang Zhou et al.

Motivated by deep neural networks, the deep Gaussian process (DGP) generalizes the standard GP by stacking multiple layers of GPs. Despite the enhanced expressiveness, GP, as an $L_2$ regularization prior, tends to be over-smooth and sub-optimal for inhomogeneous subjects, such as images with edges. Recently, Q-exponential process (Q-EP) has been proposed as an $L_q$ relaxation to GP and demonstrated with more desirable regularization properties through a parameter $q>0$ with $q=2$ corresponding to GP. Sharing the similar tractability of posterior and predictive distributions with GP, Q-EP can also be stacked to improve its modeling flexibility. In this paper, we generalize Q-EP to deep Q-EP to enjoy both proper regularization and improved expressiveness. The generalization is realized by introducing shallow Q-EP as a latent variable model and then building a hierarchy of the shallow Q-EP layers. Sparse approximation by inducing points and scalable variational strategy are applied to facilitate the inference. We demonstrate the numerical advantages of the proposed deep Q-EP model by comparing with multiple state-of-the-art deep probabilistic models.

Yiming Wang, Lingchao Guo, Zhaoming Lu et al.

In this paper, we present Wi-Mose, the first 3D moving human pose estimation system using commodity WiFi. Previous WiFi-based works have achieved 2D and 3D pose estimation. These solutions either capture poses from one perspective or construct poses of people who are at a fixed point, preventing their wide adoption in daily scenarios. To reconstruct 3D poses of people who move throughout the space rather than a fixed point, we fuse the amplitude and phase into Channel State Information (CSI) images which can provide both pose and position information. Besides, we design a neural network to extract features that are only associated with poses from CSI images and then convert the features into key-point coordinates. Experimental results show that Wi-Mose can localize key-point with 29.7mm and 37.8mm Procrustes analysis Mean Per Joint Position Error (P-MPJPE) in the Line of Sight (LoS) and Non-Line of Sight (NLoS) scenarios, respectively, achieving higher performance than the state-of-the-art method. The results indicate that Wi-Mose can capture high-precision 3D human poses throughout the space.

Shuang Zhou, Lingchao Guo, Zhaoming Lu et al.

Recently, commodity Wi-Fi devices have been shown to be able to construct human pose images, i.e., human skeletons, as fine-grained as cameras. Existing papers achieve good results when constructing the images of subjects who are in the prior training samples. However, the performance drops when it comes to new subjects, i.e., the subjects who are not in the training samples. This paper focuses on solving the subject-generalization problem in human pose image construction. To this end, we define the subject as the domain. Then we design a Domain-Independent Neural Network (DINN) to extract subject-independent features and convert them into fine-grained human pose images. We also propose a novel training method to train the DINN and it has no re-training overhead comparing with the domain-adversarial approach. We build a prototype system and experimental results demonstrate that our system can construct fine-grained human pose images of new subjects with commodity Wi-Fi in both the visible and through-wall scenarios, which shows the effectiveness and the subject-generalization ability of our model.

Sean Plummer, Shuang Zhou, Anirban Bhattacharya et al.

Transformation-based methods have been an attractive approach in non-parametric inference for problems such as unconditional and conditional density estimation due to their unique hierarchical structure that models the data as flexible transformation of a set of common latent variables. More recently, transformation-based models have been used in variational inference (VI) to construct flexible implicit families of variational distributions. However, their use in both non-parametric inference and variational inference lacks theoretical justification. We provide theoretical justification for the use of non-linear latent variable models (NL-LVMs) in non-parametric inference by showing that the support of the transformation induced prior in the space of densities is sufficiently large in the $L_1$ sense. We also show that, when a Gaussian process (GP) prior is placed on the transformation function, the posterior concentrates at the optimal rate up to a logarithmic factor. Adopting the flexibility demonstrated in the non-parametric setting, we use the NL-LVM to construct an implicit family of variational distributions, deemed GP-IVI. We delineate sufficient conditions under which GP-IVI achieves optimal risk bounds and approximates the true posterior in the sense of the Kullback-Leibler divergence. To the best of our knowledge, this is the first work on providing theoretical guarantees for implicit variational inference.

Xiang Yue, Shuang Zhou

De-identification is the task of identifying protected health information (PHI) in the clinical text. Existing neural de-identification models often fail to generalize to a new dataset. We propose a simple yet effective data augmentation method PHICON to alleviate the generalization issue. PHICON consists of PHI augmentation and Context augmentation, which creates augmented training corpora by replacing PHI entities with named-entities sampled from external sources, and by changing background context with synonym replacement or random word insertion, respectively. Experimental results on the i2b2 2006 and 2014 de-identification challenge datasets show that PHICON can help three selected de-identification models boost F1-score (by at most 8.6%) on cross-dataset test setting. We also discuss how much augmentation to use and how each augmentation method influences the performance.

Lingchao Guo, Zhaoming Lu, Shuang Zhou et al.

As elderly population grows, social and health care begin to face validation challenges, in-home monitoring is becoming a focus for professionals in the field. Governments urgently need to improve the quality of healthcare services at lower costs while ensuring the comfort and independence of the elderly. This work presents an in-home monitoring approach based on off-the-shelf WiFi, which is low-costs, non-wearable and makes all-round daily healthcare information available to caregivers. The proposed approach can capture fine-grained human pose figures even through a wall and track detailed respiration status simultaneously by off-the-shelf WiFi devices. Based on them, behavioral data, physiological data and the derived information (e.g., abnormal events and underlying diseases), of the elderly could be seen by caregivers directly. We design a series of signal processing methods and a neural network to capture human pose figures and extract respiration status curves from WiFi Channel State Information (CSI). Extensive experiments are conducted and according to the results, off-the-shelf WiFi devices are capable of capturing fine-grained human pose figures, similar to cameras, even through a wall and track accurate respiration status, thus demonstrating the effectiveness and feasibility of our approach for in-home monitoring.