arXiv subjects

Yuan Liang

Publications and source records attributed to Yuan Liang.

At least 19 recordsLinked to original sources

arXiv2026

Cast3: Translating numerical weather prediction principles into data-driven forecasting

Data-driven weather models have made rapid advances in recent years, reaching and in some metrics surpassing the large-scale forecast skill of operational numerical weather prediction. This progress, however, has been built almost entirely on the reanalysis data that NWP produced, while the methodological knowledge that the NWP community distilled over decades of multi-scale atmospheric modelling remains largely unused. Here we present Cast3, a generative forecasting framework that systematically absorbs NWP meta-knowledge to close this gap. Cast3 operates on variable-resolution cubed-sphere grids for scale-aware representation and constructs structurally diverse super-ensembles that sample the complementary biases of different grid discretizations, delivering state-of-the-art ensemble prediction. It further introduces generative nudging, a posterior-sampling strategy that distils the collective information of the full ensemble into a single forecast possessing both the large-scale accuracy of the ensemble mean and the mesoscale realism of a high-resolution member. Evaluated across synoptic-scale skill, spectral fidelity, station-level surface verification, and tropical cyclone prediction, Cast3 outperforms established deterministic and generative baselines across various dimensions. More broadly, these results demonstrate that the design principles embedded in computational atmospheric science offer a rich and largely untapped foundation for the next generation of data-driven Earth system modelling.

physics.ao-ph↗

arXiv2026

Hessian-based photometric substructure as an evolutionary tracer of OB cluster candidates in M31

Using \textit{Hubble Space Telescope} images from the PHAT and PHAST surveys, we construct an updated catalogue of 747 OB cluster (OBC) candidates. We introduce a dimensionless structural metric, the trace coefficient of variation ($CV_{\rm tr}$), derived from the Hessian matrix in four \textit{HST} bands, to quantify the internal photometric substructure of partially resolved OBC candidates. Cross-matching with the subset of M31 clusters that have independent colour--magnitude diagram (CMD) age estimates yields 247 objects in common. We find statistically significant anti-correlations between $CV_{\rm tr}$ and age in the UV and blue bands, suggesting a progressive smoothing of the light distribution as clusters evolve. Bootstrap resampling confirms the robustness of these trends. Forward modelling of synthetic clusters analysed with the same pipeline recovers a monotonic $CV_{\rm tr}$--age relation under simplified but physically motivated assumptions. These results show that second-order photometric structure contains measurable evolutionary information within the CMD-calibrated regime ($\sim10$--300~Myr).

astro-ph.GA↗

arXiv2026

A Python/CuPy Software Correlator for QUEST: Real-Time Performance and Initial Imaging

We present a Python/CuPy FX software correlator for small radio interferometer arrays and evaluate it on QUEST (Qilu University Explorer Survey Telescope). The system combines multi-threaded data ingest, pinned-memory host-device transfers, GPU-accelerated correlation, Polyphase Filter Bank channelization, MAD-based RFI flagging, and delay/phase calibration in a single workflow aimed at array commissioning. On a single NVIDIA RTX 4090D GPU, the implementation reaches a peak throughput of 1.51 GB/s, which is sufficient for real-time operation in the four-antenna mode tested here. After calibration, the visibility phase across a clean 1.32-1.38 GHz band is flattened to a residual scatter of a few degrees. Using the calibrated visibilities, we form a four-antenna synthesis image of Cassiopeia A; the CLEANed image recovers a compact source at the phase center and reduces image-domain background fluctuations from order 0.1 to a few 0.01 Jy/beam. These results indicate that the software is suitable for small-array commissioning and initial synthesis imaging on QUEST. A GNSS-based beam measurement is included as a supporting commissioning check.

astro-ph.IM↗

arXiv2026

SkillNet: Create, Evaluate, and Connect AI Skills

Current AI agents can flexibly invoke tools and execute complex tasks, yet their long-term advancement is hindered by the lack of systematic accumulation and transfer of skills. Without a unified mechanism for skill consolidation, agents frequently ``reinvent the wheel'', rediscovering solutions in isolated contexts without leveraging prior strategies. To address this challenge, we introduce SkillNet, an open infrastructure for creating, evaluating, and organizing AI skills at scale. SkillNet structures skills within a unified ontology that supports creating skills from heterogeneous sources, establishing rich relational connections, and performing multi-dimensional evaluation across Safety, Completeness, Executability, Maintainability, and Cost-awareness. Our infrastructure integrates a repository of over 600,000 skills, an interactive platform, and a versatile Python toolkit. Experiments on ALFWorld, WebShop, and ScienceWorld show 40% higher average rewards and 30% fewer execution steps across multiple backbone models. Furthermore, SkillNet-Gym benchmarks skill retrieval, utilization, and composition, while SkillNet-Fabric enables task-specific skill routing through lightweight Wikis. By formalizing skills as evolving, composable assets, SkillNet provides a robust foundation for agents to move from transient experience to durable mastery.

cs.AI↗

arXiv2025

Topology Matters: Measuring Memory Leakage in Multi-Agent LLMs

Graph topology is a fundamental determinant of memory leakage in multi-agent LLM systems, yet its effects remain poorly quantified. We introduce MAMA (Multi-Agent Memory Attack), a controlled evaluation framework for comparing topology-conditioned memory leakage in multi-agent LLM systems. MAMA operates on synthetic documents containing labeled Personally Identifiable Information (PII) entities, from which we generate sanitized task instructions. We execute a two-phase protocol: Engram (seeding private information into a target agent's memory) and Resonance (multi-round interaction where an attacker attempts extraction). Over 10 rounds, we measure leakage using a two-stage recovery criterion that combines exact-match extraction with LLM-based inference over the attacker's final output. We evaluate six canonical topologies (complete, circle, chain, tree, star, star-ring) across $n\in\{4,5,6\}$, attacker-target placements, and base models. Results are consistent: denser connectivity, shorter attacker-target distance, and higher target centrality increase leakage; most leakage occurs in early rounds and then plateaus; model choice shifts absolute rates but preserves broad structural trends; spatiotemporal/location attributes leak more readily than identity credentials or regulated identifiers. We distill practical guidance for system design: favor sparse or hierarchical connectivity, maximize attacker-target separation, and restrict hub/shortcut pathways via topology-aware access control. Our code is available at https://github.com/llll121/mama-eval.

cs.CR↗

arXiv2025

O-Mem: Omni Memory System for Personalized, Long Horizon, Self-Evolving Agents

Recent advancements in LLM-powered agents have demonstrated significant potential in generating human-like responses; however, they continue to face challenges in maintaining long-term interactions within complex environments, primarily due to limitations in contextual consistency and dynamic personalization. Existing memory systems often depend on semantic grouping prior to retrieval, which can overlook semantically irrelevant yet critical user information and introduce retrieval noise. In this report, we propose the initial design of O-Mem, a novel memory framework based on active user profiling that dynamically extracts and updates user characteristics and event records from their proactive interactions with agents. O-Mem supports hierarchical retrieval of persona attributes and topic-related context, enabling more adaptive and coherent personalized responses. O-Mem achieves 51.67% on the public LoCoMo benchmark, a nearly 3% improvement upon LangMem,the previous state-of-the-art, and it achieves 62.99% on PERSONAMEM, a 3.5% improvement upon A-Mem,the previous state-of-the-art. O-Mem also boosts token and interaction response time efficiency compared to previous memory frameworks. Our work opens up promising directions for developing efficient and human-like personalized AI assistants in the future.

cs.CL↗

arXiv2025

ASTROFLOW: A Real-Time End-to-End Pipeline for Radio Single-Pulse Searches

Fast radio bursts (FRBs) are extremely bright, millisecond duration cosmic transients of unknown origin. The growing number of wide-field and high-time-resolution radio surveys, particularly with next-generation facilities such as the SKA and MeerKAT, will dramatically increase FRB discovery rates, but also produce data volumes that overwhelm conventional search pipelines. Real-time detection thus demands software that is both algorithmically robust and computationally efficient. We present Astroflow, an end-to-end, GPU-accelerated pipeline for single-pulse detection in radio time-frequency data. Built on a unified C++/CUDA core with a Python interface, Astroflow integrates RFI excision, incoherent dedispersion, dynamic-spectrum tiling, and a YOLO-based deep detector. Through vectorized memory access, shared-memory tiling, and OpenMP parallelism, it achieves 10x faster-than-real-time processing on consumer GPUs for a typical 150 s, 2048-channel observation, while preserving high sensitivity across a wide range of pulse widths and dispersion measures. These results establish the feasibility of a fully integrated, GPU-accelerated single-pulse search stack, capable of scaling to the data volumes expected from upcoming large-scale surveys. Astroflow offers a reusable and deployable solution for real-time transient discovery, and provides a framework that can be continuously refined with new data and models.

astro-ph.IM↗

arXiv2025

Towards Personalized Deep Research: Benchmarks and Evaluations

Deep Research Agents (DRAs) can autonomously conduct complex investigations and generate comprehensive reports, demonstrating strong real-world potential. However, existing evaluations mostly rely on close-ended benchmarks, while open-ended deep research benchmarks remain scarce and typically neglect personalized scenarios. To bridge this gap, we introduce Personalized Deep Research Bench (PDR-Bench), the first benchmark for evaluating personalization in DRAs. It pairs 50 diverse research tasks across 10 domains with 25 authentic user profiles that combine structured persona attributes with dynamic real-world contexts, yielding 250 realistic user-task queries. To assess system performance, we propose the PQR Evaluation Framework, which jointly measures Personalization Alignment, Content Quality, and Factual Reliability. Our experiments on a range of systems highlight current capabilities and limitations in handling personalized deep research. This work establishes a rigorous foundation for developing and evaluating the next generation of truly personalized AI research assistants.

cs.CL↗

arXiv2025

Improving LLMs' Learning for Coreference Resolution

Coreference Resolution (CR) is crucial for many NLP tasks, but existing LLMs struggle with hallucination and under-performance. In this paper, we investigate the limitations of existing LLM-based approaches to CR-specifically the Question-Answering (QA) Template and Document Template methods and propose two novel techniques: Reversed Training with Joint Inference and Iterative Document Generation. Our experiments show that Reversed Training improves the QA Template method, while Iterative Document Generation eliminates hallucinations in the generated source text and boosts coreference resolution. Integrating these methods and techniques offers an effective and robust solution to LLM-based coreference resolution.

cs.CL↗

arXiv2025

A comprehensive catalogue of OB cluster candidates in M31 and their association with giant molecular clouds

We present a new catalogue of 578 OB cluster (OBC) candidates in the Andromeda galaxy (M31), identified using a MeanShift-based algorithm on HST's F275W-band imaging from the PHAT (Dalcanton et al. 2012) and PHAST (Chen et al. 2025) Hubble surveys. These clusters exhibit typical half-light radii of 1-2 pc and strong ultraviolet luminosities indicative of recent massive star formation. Spatial analysis reveals a pronounced north-south asymmetry: clusters in the northern disc show tight associations with giant molecular clouds (GMCs), while southern clusters appear more compact and luminous but less correlated with molecular gas. Two-point correlation functions demonstrate significant clustering of OBC candidates on scales 100 pc and a strong spatial association with GMCs, consistent with hierarchical star formation in dense gas-rich environments. These findings offer new constraints on the early evolution and feedback-driven dispersal of young stellar clusters across galactic discs.

astro-ph.GA↗

arXiv2025

NeuroDx-LM: A Clinical Large-Scale Model for EEG-based Neurological Disorder Detection

Large-scale models pre-trained on Electroencephalography (EEG) have shown promise in clinical applications such as neurological disorder detection. However, the practical deployment of EEG-based large-scale models faces critical challenges such as limited labeled EEG data and suboptimal performance in clinical scenarios. To address these issues, we propose NeuroDx-LM, a novel large-scale model specifically designed for detecting EEG-based neurological disorders. Our key contributions include (i) a Selective Temporal-Frequency Embedding mechanism that adaptively captures complex temporal and spectral patterns in EEG signals; and (ii) a Progressive Feature-Aware Training strategy that refines feature representation in a two-stage process. In the first stage, our model learns the fundamental discriminative features of EEG activities; in the second stage, the model further extracts more specialized fine-grained features for accurate diagnostic performance. We evaluated NeuroDx-LM on the CHB-MIT and Schizophrenia datasets, achieving state-of-the-art performance in EEG-based seizure and schizophrenia detection, respectively. These results demonstrate the great potential of EEG-based large-scale models to advance clinical applicability. Our code is available at https://github.com/LetItBe12345/NeuroDx-LM.

cs.LG↗

arXiv2025

Memp: Exploring Agent Procedural Memory

Large Language Models (LLMs) based agents excel at diverse tasks, yet they suffer from brittle procedural memory that is manually engineered or entangled in static parameters. In this work, we investigate strategies to endow agents with a learnable, updatable, and lifelong procedural memory. We propose Memp that distills past agent trajectories into both fine-grained, step-by-step instructions and higher-level, script-like abstractions, and explore the impact of different strategies for Build, Retrieval, and Update of procedural memory. Coupled with a dynamic regimen that continuously updates, corrects, and deprecates its contents, this repository evolves in lockstep with new experience. Empirical evaluation on TravelPlanner and ALFWorld shows that as the memory repository is refined, agents achieve steadily higher success rates and greater efficiency on analogous tasks. Moreover, procedural memory built from a stronger model retains its value: migrating the procedural memory to a weaker model can also yield substantial performance gains. Code is available at https://github.com/zjunlp/MemP.

cs.CL↗

arXiv2025

FRAM: Frobenius-Regularized Assignment Matching with Mixed-Precision Computing

Graph matching, typically formulated as a Quadratic Assignment Problem (QAP), seeks to establish node correspondences between two graphs. To address the NP-hardness of QAP, some existing methods adopt projection-based relaxations that embed the problem into the convex hull of the discrete domain. However, these relaxations inevitably enlarge the feasible set, introducing two sources of error: numerical scale sensitivity and geometric misalignment between the relaxed and original domains. To alleviate these errors, we propose a novel relaxation framework by reformulating the projection step as a Frobenius-regularized Linear Assignment (FRA) problem, where a tunable regularization term mitigates feasible region inflation. This formulation enables normalization-based operations to preserve numerical scale invariance without compromising accuracy. To efficiently solve FRA, we propose the Scaling Doubly Stochastic Normalization (SDSN) algorithm. Building on its favorable computational properties, we develop a theoretically grounded mixed-precision architecture to achieve substantial acceleration. Comprehensive CPU-based benchmarks demonstrate that FRAM consistently outperforms all baseline methods under identical precision settings. When combined with a GPU-based mixed-precision architecture, FRAM achieves up to 370X speedup over its CPU-FP64 counterpart, with negligible loss in solution accuracy.

cs.LG↗

arXiv2025

Fusing Radiomic Features with Deep Representations for Gestational Age Estimation in Fetal Ultrasound Images

Accurate gestational age (GA) estimation, ideally through fetal ultrasound measurement, is a crucial aspect of providing excellent antenatal care. However, deriving GA from manual fetal biometric measurements depends on the operator and is time-consuming. Hence, automatic computer-assisted methods are demanded in clinical practice. In this paper, we present a novel feature fusion framework to estimate GA using fetal ultrasound images without any measurement information. We adopt a deep learning model to extract deep representations from ultrasound images. We extract radiomic features to reveal patterns and characteristics of fetal brain growth. To harness the interpretability of radiomics in medical imaging analysis, we estimate GA by fusing radiomic features and deep representations. Our framework estimates GA with a mean absolute error of 8.0 days across three trimesters, outperforming current machine learning-based methods at these gestational ages. Experimental results demonstrate the robustness of our framework across different populations in diverse geographical regions. Our code is publicly available on \href{https://github.com/13204942/RadiomicsImageFusion_FetalUS}.

eess.IV↗

arXiv2025

SynWorld: Virtual Scenario Synthesis for Agentic Action Knowledge Refinement

In the interaction between agents and their environments, agents expand their capabilities by planning and executing actions. However, LLM-based agents face substantial challenges when deployed in novel environments or required to navigate unconventional action spaces. To empower agents to autonomously explore environments, optimize workflows, and enhance their understanding of actions, we propose SynWorld, a framework that allows agents to synthesize possible scenarios with multi-step action invocation within the action space and perform Monte Carlo Tree Search (MCTS) exploration to effectively refine their action knowledge in the current environment. Our experiments demonstrate that SynWorld is an effective and general approach to learning action knowledge in new environments. Code is available at https://github.com/zjunlp/SynWorld.

cs.CL↗

arXiv2025

TripoSG: High-Fidelity 3D Shape Synthesis using Large-Scale Rectified Flow Models

Recent advancements in diffusion techniques have propelled image and video generation to unprecedented levels of quality, significantly accelerating the deployment and application of generative AI. However, 3D shape generation technology has so far lagged behind, constrained by limitations in 3D data scale, complexity of 3D data processing, and insufficient exploration of advanced techniques in the 3D domain. Current approaches to 3D shape generation face substantial challenges in terms of output quality, generalization capability, and alignment with input conditions. We present TripoSG, a new streamlined shape diffusion paradigm capable of generating high-fidelity 3D meshes with precise correspondence to input images. Specifically, we propose: 1) A large-scale rectified flow transformer for 3D shape generation, achieving state-of-the-art fidelity through training on extensive, high-quality data. 2) A hybrid supervised training strategy combining SDF, normal, and eikonal losses for 3D VAE, achieving high-quality 3D reconstruction performance. 3) A data processing pipeline to generate 2 million high-quality 3D samples, highlighting the crucial rules for data quality and quantity in training 3D generative models. Through comprehensive experiments, we have validated the effectiveness of each component in our new framework. The seamless integration of these parts has enabled TripoSG to achieve state-of-the-art performance in 3D shape generation. The resulting 3D shapes exhibit enhanced detail due to high-resolution capabilities and demonstrate exceptional fidelity to input images. Moreover, TripoSG demonstrates improved versatility in generating 3D models from diverse image styles and contents, showcasing strong generalization capabilities. To foster progress and innovation in the field of 3D generation, we will make our model publicly available.

cs.CV↗

arXiv2025

ALATDYN: a set of Anharmonic LATtice DYNamics codes to compute thermodynamic and thermal transport properties of crystalline solids

We introduce a lattice dynamics package which calculates elastic, thermodynamic and thermal transport properties of crystalline materials from data on their force and potential energy as a function of atomic positions. The data can come from density functional theory (DFT) calculations or classical molecular dynamics runs performed in a supercell. First, the model potential parameters, which are anharmonic force constants are extracted from the latter runs. Then, once the anharmonic model is defined, thermal conductivity and equilibrium properties at finite temperatures can be computed using lattice dynamics, Boltzmann transport theories, and a variational principle respectively. In addition, the software calculates the mechanical properties such as elastic tensor, Gruneisen parameters and the thermal expansion coefficient within the quasi-harmonic approximation (QHA). Phonons, elastic constants and thermodynamic properties results applied to the germanium crystal will be illustrated. Using the force constants as a force field, one may also perform molecular dynamics (MD) simulations in order to investigate the combined effects of anharmonicity and defect scattering beyond perturbation theory.

cond-mat.mtrl-sci↗

arXiv2024

Boosting weather forecast via generative superensemble

Accurate weather forecasting is essential for socioeconomic activities. While data-driven forecasting demonstrates superior predictive capabilities over traditional Numerical Weather Prediction (NWP) with reduced computational demands, its deterministic nature and limited advantages over physics-based ensemble predictions restrict operational applications. We introduce the generative ensemble prediction system (GenEPS) framework to address these limitations by randomizing and mitigating both random errors and systematic biases. GenEPS provides a plug-and-play ensemble forecasting capability for deterministic models to eliminate random errors, while incorporating cross-model integration for cross-model ensembles to address systematic biases. The framework culminates in a super-ensemble approach utilizing all available data-driven models to further minimize systematic biases. GenEPS achieves an Anomaly Correlation Coefficient (ACC) of 0.679 for 500hPa geopotential (Z500), exceeding the ECMWF Ensemble Prediction System's (ENS) ACC of 0.646. Integration of the ECMWF ensemble mean further improves the ACC to 0.683. The framework also enhances extreme event representation and produces energy spectra more consistent with ERA5 reanalysis. GenEPS establishes a new paradigm in ensemble forecasting by enabling the integration of multiple data-driven models into a high-performing super-ensemble system.

physics.ao-ph↗