Xu Xu

Jiangyuan Wang, Xuyong Chen, Junwei He et al.

Long-horizon clinical simulation -- predicting how a patient's physiology evolves over years under specified interventions -- is central to chronic-disease care, yet existing electronic health record (EHR) models are predominantly discriminative, and general-purpose large language models drift under repeated interventions. We propose the \textbf{ChronoMedicalWorld Model (CMWM)}, an action-conditioned latent world-model framework for learning patient trajectories from longitudinal care data. CMWM couples a joint-embedding state encoder with a wide action encoder that admits both structured intervention indicators and free-text communication embeddings, and trains a recurrent latent transition module under a six-term objective: next-observation supervision, next-latent prediction, SIGReg latent regularisation, and three physiology-aware shape priors (slope, continuity, large-jump penalty). A closed-loop rollout-prefix protocol matches training to deployment, so the model is optimised against the same multi-step error it exhibits at inference. As a concrete case study, we instantiate CMWM for annual estimated glomerular filtration rate (eGFR) trajectory forecasting in chronic kidney disease (CKD). On a 2{,}232-patient nephrology cohort, the CKD instantiation achieves a dynamic-50\% history rollout test mean absolute error (MAE) of 7.384 and root-mean-square error (RMSE) of 10.256, against 7.964 and 11.069 for a tuned GPT-5.5 structured-prompting baseline ($-7.28\%$ MAE, $-7.35\%$ RMSE), with the gain dominated by the dialogue portion of patient--health-coach communication. The framework is not CKD-specific: its architecture, loss design, and training protocol apply to any chronic condition that can be cast as periodic clinical state interleaved with structured and conversational interventions.

Wenzhen Zheng, Wenbo Pan, Xu Xu et al.

In recent years, Large Language Models (LLMs) have made significant strides towards Artificial General Intelligence. However, training these models from scratch requires substantial computational resources and vast amounts of text data. In this paper, we explore an alternative approach to constructing an LLM for a new language by continually pretraining (CPT) from existing pretrained LLMs, instead of using randomly initialized parameters. Based on parallel experiments on 40 model sizes ranging from 40M to 5B parameters, we find that 1) CPT converges faster and saves significant resources in a scalable manner; 2) CPT adheres to an extended scaling law derived from Hoffmann et al. (2022) with a joint data-parameter scaling term; 3) The compute-optimal data-parameter allocation for CPT markedly differs based on our estimated scaling factors; 4) The effectiveness of transfer at scale is influenced by training duration and linguistic properties, while robust to data replaying, a method that effectively mitigates catastrophic forgetting in CPT. We hope our findings provide deeper insights into the transferability of LLMs at scale for the research community.

Haoran Sun, Haoyu Bian, Shaoning Zeng et al.

Common knowledge indicates that the process of constructing image datasets usually depends on the time-intensive and inefficient method of manual collection and annotation. Large models offer a solution via data generation. Nonetheless, real-world data are obviously more valuable comparing to artificially intelligence generated data, particularly in constructing image datasets. For this reason, we propose a novel method for auto-constructing datasets from real-world images by a multiagent collaborative system, named as DatasetAgent. By coordinating four different agents equipped with Multi-modal Large Language Models (MLLMs), as well as a tool package for image optimization, DatasetAgent is able to construct high-quality image datasets according to user-specified requirements. In particular, two types of experiments are conducted, including expanding existing datasets and creating new ones from scratch, on a variety of open-source datasets. In both cases, multiple image datasets constructed by DatasetAgent are used to train various vision models for image classification, object detection, and image segmentation.

You Peng, Youhe Jiang, Wenshuang Li et al.

Agentic LLM applications increasingly execute user requests as multi-step workflows involving planning, tool use, branching, refinement, and synthesis. In such settings, users experience the end-to-end latency of an entire workflow, not the latency of any single LLM call. In this paper, we study how to schedule online agentic workflows across heterogeneous prefill-decode disaggregated LLM serving clusters to efficiently meet workflow-level latency objectives. The problem is challenging because workflow dependencies are revealed incrementally at runtime, calls have heterogeneous prompts, outputs, and KV-cache requirements, and the prefill and decode stages impose different compute, memory, and transfer constraints across heterogeneous GPUs. To solve this problem, we present HexAGenT, a workflow-aware scheduler for a heterogeneous prefill-decode inference service. HexAGenT models each request as an online-revealed DAG, maintains a running estimate of the workflow's standalone completion horizon, prioritizes ready calls by projected risk of missing that horizon, and jointly selects prefill placement, decode placement, and local queue priority while accounting for KV-cache capacity and cross-stage transfer latency. Across representative agentic workloads and heterogeneous A100/H100/H200 clusters, HexAGenT reduces the SLO scale required for timely workflow completion by an average of 20.1% at 95% attainment and 33.0% at 99% attainment, with maximum reductions of 45.0% and 80.5%, respectively.

Chuanhao Yan, Fengdi Che, Xuhan Huang et al.

Existing informal language-based (e.g., human language) Large Language Models (LLMs) trained with Reinforcement Learning (RL) face a significant challenge: their verification processes, which provide crucial training signals, are neither reliable nor scalable. In fact, the prevalent large proprietary models could hardly generate verifiable programs. A promising yet largely uncharted alternative is formal language-based reasoning. Grounding LLMs in rigorous formal systems where generative models operate in formal language spaces (e.g., Dafny) enables the automatic and mathematically provable verification of their reasoning processes and outcomes. This capability is pivotal for achieving large-scale, reliable formal software verification. It is a common practice to employ human-annotated chain-of-thought and other human priors to induce the reasoning and coding capabilities of LLMs. Unfortunately, it becomes unacceptably all-consuming to provide such priors for supervising complex programming tasks. In this work, we systematically explore ways to reduce human priors with the formal language, Dafny, as the main environment for our pilot study. Our pipeline mainly relies on introducing an automatic and scalable data curation pipeline, and careful RL designs integrated with feedback from the formal language verifier. We introduce DafnyComp, a benchmark of compositional formal programs with auto-formalized specifications for specification reasoning. Our supervised fine-tuning (SFT) stage enables even small models (e.g., 0.5B) to generate syntactically valid and verifiable Dafny code, surpassing proprietary models. RL with regularization further improves performance, achieving stronger generalization to out-of-domain tasks and outperforming all strong baselines on the challenging DafnyComp benchmark.

Lingfei Zeng, Fengdi Che, Xuhan Huang et al.

Formal verification is the next frontier for ensuring the correctness of code generated by Large Language Models (LLMs). While methods that co-generate code and formal specifications in formal languages, like Dafny, can, in principle, prove alignment with user intent, progress is bottlenecked by specification quality evaluation. Current benchmarks rely on matching against ground-truth specifications, a manual and expertise-intensive process that has limited existing datasets to a few hundred simple problems and also suffers from a reliability issue. To address this, we introduce VeriEquivBench, a new benchmark with $2,389$ complex algorithmic problems that probe the limitations of current models in both code generation and formal reasoning. Our evaluation framework replaces ground-truth matching with a formally grounded metric, the equivalence score, and rigorously verifies the quality of generated specifications and code. Our results show that generating formally verifiable code remains a profound challenge for state-of-the-art LLMs. This underscores both the difficulty of the task and the need for benchmarks like VeriEquivBench to drive progress toward scalable and reliable coding agents.

Xu Xu, Xin Li, Xingwei Qu et al.

We introduce DafnyCOMP, a benchmark for evaluating large language models (LLMs) on compositional specification generation in Dafny. Unlike prior benchmarks that focus on single-function tasks, DafnyCOMP targets programs composed of multiple interacting functions with data dependencies, requiring reasoning across component boundaries. The benchmark consists of 300 automatically synthesized multi-function programs. We evaluate several state-of-the-art LLM families and find that, while they perform well on single-function verification, their performance drops sharply on compositional tasks. Analysis reveals systematic failures in cross-functional reasoning, including fragile specifications, misalignment between implementations and proofs, and unstable reasoning. DafnyCOMP thus provides a diagnostic tool for measuring progress toward reliable, verifiable, and compositional code generation with LLMs.

Chunming Ye, Songzhou Li, Xu Xu

The performance of Large Language Models is influenced by their characteristics such as architecture, model sizes, decoding methods and so on. Due to differences in structure or function, the weights in different layers of large models have varying distributions. This paper explores the correlations between different types of layers in terms of weights distribution and studies the potential impact of these correlations on LoRA training effectiveness. Firstly, the study reveals that in the model the cosine distances between weights of different layers manifest power-law distribution. We extract Query-projection, down-projection and other weight matrices from the self-attention layers and MLP layers, calculate the singular values of the matrices using singular value decomposition, and organize a certain number of singular values into matrices according to projection's type. By analyzing the probability distribution of the cosine distances between these matrices, it is found that the cosine distances values between them have distinct power-law distribution characteristics. Secondly, based on the results of distance calculations and analysis across different layers of model, a qualitative method is proposed to describe the distribution characteristics of different models. Next, to construct weights that align with the distribution characteristics, a data generator is designed using a combination of Gaussian process and Pareto distribution functions. The generator is used to simulate the generation of data that aligns with specific distribution characteristics. Finally, based on the aforementioned distribution characteristics and data generation method, the weights in LoRA initialization are reshaped for training. Experimental results indicate that, without altering the model structure or training process, this method achieves a certain improvement in the performance of LoRA training.

Johann Li, Guangming Zhu, Cong Hua et al.

The astounding success made by artificial intelligence (AI) in healthcare and other fields proves that AI can achieve human-like performance. However, success always comes with challenges. Deep learning algorithms are data-dependent and require large datasets for training. The lack of data in the medical imaging field creates a bottleneck for the application of deep learning to medical image analysis. Medical image acquisition, annotation, and analysis are costly, and their usage is constrained by ethical restrictions. They also require many resources, such as human expertise and funding. That makes it difficult for non-medical researchers to have access to useful and large medical data. Thus, as comprehensive as possible, this paper provides a collection of medical image datasets with their associated challenges for deep learning research. We have collected information of around three hundred datasets and challenges mainly reported between 2013 and 2020 and categorized them into four categories: head & neck, chest & abdomen, pathology & blood, and ``others''. Our paper has three purposes: 1) to provide a most up to date and complete list that can be used as a universal reference to easily find the datasets for clinical image analysis, 2) to guide researchers on the methodology to test and evaluate their methods' performance and robustness on relevant datasets, 3) to provide a ``route'' to relevant algorithms for the relevant medical topics, and challenge leaderboards.

Rahil Mehrizi, Xi Peng, Zhiqiang Tang et al.

Lifting is a common manual material handling task performed in the workplaces. It is considered as one of the main risk factors for Work-related Musculoskeletal Disorders. To improve work place safety, it is necessary to assess musculoskeletal and biomechanical risk exposures associated with these tasks, which requires very accurate 3D pose. Existing approaches mainly utilize marker-based sensors to collect 3D information. However, these methods are usually expensive to setup, time-consuming in process, and sensitive to the surrounding environment. In this study, we propose a multi-view based deep perceptron approach to address aforementioned limitations. Our approach consists of two modules: a "view-specific perceptron" network extracts rich information independently from the image of view, which includes both 2D shape and hierarchical texture information; while a "multi-view integration" network synthesizes information from all available views to predict accurate 3D pose. To fully evaluate our approach, we carried out comprehensive experiments to compare different variants of our design. The results prove that our approach achieves comparable performance with former marker-based methods, i.e. an average error of $14.72 \pm 2.96$ mm on the lifting dataset. The results are also compared with state-of-the-art methods on HumanEva-I dataset, which demonstrates the superior performance of our approach.

Xu Xu, Sinisa Todorovic

This paper addresses 3D shape recognition. Recent work typically represents a 3D shape as a set of binary variables corresponding to 3D voxels of a uniform 3D grid centered on the shape, and resorts to deep convolutional neural networks(CNNs) for modeling these binary variables. Robust learning of such CNNs is currently limited by the small datasets of 3D shapes available, an order of magnitude smaller than other common datasets in computer vision. Related work typically deals with the small training datasets using a number of ad hoc, hand-tuning strategies. To address this issue, we formulate CNN learning as a beam search aimed at identifying an optimal CNN architecture, namely, the number of layers, nodes, and their connectivity in the network, as well as estimating parameters of such an optimal CNN. Each state of the beam search corresponds to a candidate CNN. Two types of actions are defined to add new convolutional filters or new convolutional layers to a parent CNN, and thus transition to children states. The utility function of each action is efficiently computed by transferring parameter values of the parent CNN to its children, thereby enabling an efficient beam search. Our experimental evaluation on the 3D ModelNet dataset demonstrates that our model pursuit using the beam search yields a CNN with superior performance on 3D shape classification than the state of the art.