arXiv subjects

Wei Chen

Publications and source records attributed to Wei Chen.

At least 181 records · Page 10Linked to original sources

arXiv2026

Study of the $Ω_{ccc}Ω_{ccc}$ and $Ω_{bbb}Ω_{bbb}$ dibaryons in QCD Sum Rules

The recent observation of a family of fully-charm tetraquark states by the LHCb, ATLAS and CMS Collaborations suggests the possible existence of fully-heavy dibaryons. In this work, we investigate the $Ω_{ccc}Ω_{ccc}$ and $Ω_{bbb}Ω_{bbb}$ dibaryons in both the $^1S_0$ and $^5S_2$ channels using the method of QCD sum rules. We employ the iterative dispersion relation (IDR) method to efficiently compute the massive five-loop banana diagrams that appear in these systems, and properly address the tricky small-circle divergence problem in the nonperturbative terms. Our analyses reveal that for both charm and bottom systems, the scalar dibaryon lies lower than its tensor counterpart. In $\overline{\text{MS}}$ scheme, the mass of the scalar $Ω_{ccc}Ω_{ccc}$ dibaryon is found to be slightly above the $2Ω_{ccc}$ mass threshold, while the $Ω_{bbb}Ω_{bbb}$ systems may form bound states. However, they are predicted to be much heavier in the on-shell scheme.

hep-ph↗

arXiv2026

ProSPy: A Profiling-Driven SQL-Python Agentic Framework for Enterprise Text-to-SQL

Large language models have substantially advanced Text-to-SQL systems, yet applying them to enterprise-scale databases remains challenging. Real-world databases often contain large and heterogeneous schemas, incomplete metadata, dialect-specific SQL syntax, and complex analytical questions that are difficult to solve with a single SQL query. To address these challenges, we propose ProSPy, a Profiling-driven SQL--Python agentic framework for enterprise-scale Text-to-SQL. ProSPy structures the reasoning process into four stages: it first extracts fine-grained data evidence through automatic profiling, progressively prunes large schemas into task-relevant contexts, fetches intermediate views through a dialect-agnostic SQL interface, and finally performs flexible downstream analysis with Python. This design combines the efficiency of SQL over large databases with the flexibility of Python-based analysis, while reducing reliance on unreliable metadata and improving robustness across SQL dialects. Experiments on Spider 2.0-Lite and Spider 2.0-Snow show that ProSPy consistently outperforms strong baselines with both open-source and proprietary models, achieving execution accuracies of 60.15% and 60.51% with Claude-4.5-Opus, without majority voting. Further analysis shows that ProSPy is robust to SQL dialect variations and achieves a favorable trade-off between schema recall and precision.

cs.CL↗

arXiv2026

StageFrontier: Synchronization-Aware Stage Accounting for Distributed ML Training

When a distributed training job slows down, the hard part is knowing where to look. Synchronization hides the cause: a stall on one rank shows up as a wait on the others, so a data delay on a single rank can surface as backward time across the group. The cheap dashboards that run all the time -- per-stage averages and maxima -- misread this, double-counting the same exposed delay or burying the slow rank in an average, while full profilers see it clearly but are far too heavy to leave on. StageFrontier is an always-on signal that closes this gap. Each rank reports only a short ordered vector of coarse stage durations -- data, forward, backward, and so on -- timed with CPU wall-clock, with no synchronized clocks and no kernel tracing. At each stage boundary, StageFrontier takes the cumulative time of whichever rank is furthest along; the increments of this frontier form an exact, additive accounting of the step's exposed time and point to the stage and rank where group-visible delay first appears, telling an operator where to aim a heavy profiler, not which fix to make. The accounting is exact, but the coarse signal alone cannot tell whether a leading stage truly caused the slowdown or merely ran alongside it; StageFrontier labels the windows where that distinction needs more evidence instead of guessing. A PyTorch implementation adds under 0.2% throughput overhead through 128 ranks on Gloo and NCCL, places injected faults among its top two suspects on all 50 rows of a hidden-rank DDP test, and recovers the same top-stage routing as PyTorch Profiler, HTA, and Nsight Systems once their traces are reduced to the same coarse stages -- from a 0.11 MB summary instead of a 15.81 GB trace.

cs.DC↗

arXiv2026

LifeSide: Benchmarking Agents as Lifelong Digital Companions

Lifelong digital companions must integrate cross-session cues, continually update their understanding of users, and adapt to shifting privacy boundaries. Existing evaluations fail to capture this, testing memory recall and short-term empathy in isolation. To bridge this gap, we introduce \benchmark, a benchmark centered on multi-session \textit{Memory-Emotion-Environment} loops. By modeling users as persistent worlds with layered profiles and event trajectories, \benchmark uses multi-agent simulation to project environmental dynamics into dialogue, preserving the critical gap between latent thoughts and observable expressions. Evaluating 2,000 personas and 111K tasks across memory tracking, user understanding, privacy control, and emotional companionship, our experiment results reveal a stark reality: even models that saturate current memory benchmarks fail to sustain accurate user understanding and true companionship over long horizons.

cs.CL↗

arXiv2026

A Visual Analytics System for Interactive Exploration of Historical Painter Cohorts

Painter cohort analysis has long been regarded as a key lens for studying how painting artistic styles develop and transmit across generations. Through a two-year collaboration with art historians, we identify key challenges in traditional painter cohort research: the unstructured characteristic of painter features, the entangled complexity of inheritance relationships, and the cognitively demanding nature of cohort definition and validation. To solve these challenges, we propose HPC-Vis, a visual analytics system for interactive exploration of historical painter cohorts. An improved cohort analytical workflow is designed to integrate structured feature construction, visualization-assisted exploration, algorithm-based recommendation, and unified cohort management. Based on this workflow, we develop three core computational modules: a multi-scale artistic feature construction method that leverages LLMs to extract and organize hierarchical style features from unstructured historical texts, an inheritance reconstruction algorithm that transforms the entangled multi-parent inheritance network into a clear hierarchical forest structure, and a recommendation model that identifies core features of the cohort and recommends cohort members via painter relevance assessment. To support smooth interactive exploration, we further design a set of novel visualizations with multidimensional collaboration, especially an inheriting mountain view inspired by traditional Chinese landscape paintings, and a foldable doughnut chart for hierarchical artistic style labels. HPC-Vis is evaluated and validated through case studies, user studies, and technical evaluations, demonstrating its effectiveness in supporting painter cohort exploration and in providing visual insights for art historical research.

cs.HC↗

arXiv2026

Hybrid-plasticity Photonic Synapses Enabling Hardware-Level Neural Reuse

Biological intelligence is distinguished by neural reuse, the capacity to preserve established learning memory while repurposing it for new tasks and dynamic environments. Bringing this capability to photonic hardware requires hybrid plasticity, namely the coexistence of long-term synaptic plasticity for persistent weight storage and short-term synaptic plasticity for rapid, reversible adaptation within a single synaptic element; however, current photonic architectures lack such a unified mechanism. Here, we demonstrate a hybrid-plasticity photonic synapse on thin-film lead zirconate titanate (PZT) that couples non-volatile and volatile modes to enable hardware-level neural reuse. Crucially, high-speed refresh operations can be superimposed without perturbing the stored weight. Such a neural-reuse framework yields a convergence speedup of over 20-fold and reduces the weight updates by approximately 30-fold compared with random initialization. These results establish hybrid-plasticity photonic synapses as a pathway toward on-chip learning systems that are both memory-preserving and rapidly adaptable.

physics.optics↗

arXiv2026

Message Tuning Outshines Graph Prompt Tuning: A Prismatic Space Perspective

Graph Foundation Models (GFMs), built upon the Pre-training and Adaptation paradigm, have emerged as a research hotspot in graph learning. For GNN-based GFMs, graph prompt tuning has become the prevailing adaptation method for downstream tasks. Although recent methods explain why graph prompt tuning works, how to rigorously measure its adaptation capacity remains an open problem. Addressing this problem is critical for understanding the capability limits of graph prompt tuning and for developing more powerful adaptation methods. In this paper, we propose Prismatic Space Theory (PS-Theory), a novel mathematical framework to quantify the capacity of adaptation methods, while focusing on establishing the upper bound for the adaptation capacity of graph prompt tuning. Building upon the proposed PS-Theory, we further introduce Message Tuning for GFMs (MTG), a lightweight approach that injects a small set of learnable message prototypes into each layer of the GNN backbone to adaptively guide message fusion without updating pre-trained weights. Through our PS-Theory, we prove that the adaptation capacity of MTG can exceed the theoretical upper bound of graph prompt tuning. Extensive experiments demonstrate that MTG consistently outperforms graph prompt baselines across diverse benchmark datasets, providing strong empirical support for our theoretical findings.

cs.LG↗

arXiv2026

Ultrafast and long-term stability Integrated Pockels laser with thin-film PZT

Integrated tunable lasers are central to coherent communications, wavelength-routed optical interconnects, spectroscopy and frequency-modulated continuous-wave LiDAR, yet chip-scale sources rarely combine broad wavelength coverage, nanosecond switching, high spectral purity and stable high-power operation. Here we demonstrate a frequency-agile hybrid external-cavity laser enabled by the Pockels effect in thin-film lead zirconate titanate (PZT). The strong linear electro-optic response of PZT provides direct, non-thermal tuning of compact microring resonators with a wavelength-tuning efficiency of 17 pm/V. In contrast to conventional anisotropic Pockels materials, the near-isotropic in-plane electro-optic behaviour of thin-film PZT relaxes crystal-axis layout constraints, allowing efficient Vernier wavelength selection in compact ring cavities. The PZT resonators also show no measurable photorefractive resonance distortion and no resolvable DC-bias drift during operation, preserving stable wavelength-selective feedback. The demonstrated laser achieves an 82 nm tuning range, a 5 mW fiber-coupled output power, a side-mode suppression ratio (SMSR) exceeding 56.7 dB, and a wavelength-switching time of 5.5 ns. These results establish thin-film PZT photonics as a powerful electro-optic platform for compact, high-power, and frequency-agile integrated laser sources.

physics.optics↗

arXiv2026

BERT4beam: Large AI Model Enabled Generalized Beamforming Optimization

Artificial intelligence (AI) is anticipated to emerge as a pivotal enabler for the forthcoming sixth-generation (6G) wireless communication systems. However, current research efforts regarding large AI models for wireless communications primarily focus on fine-tuning pre-trained large language models (LLMs) for specific tasks. This paper investigates the large-scale AI model designed for beamforming optimization to adapt and generalize to diverse tasks defined by system utilities and scales. We propose a novel framework based on bidirectional encoder representations from transformers (BERT), termed BERT4beam. We aim to formulate the beamforming optimization problem as a token-level sequence learning task, perform tokenization of the channel state information, construct the BERT model, and conduct task-specific pre-training and fine-tuning strategies. Based on the framework, we propose two BERT-based approaches for single-task and multi-task beamforming optimization, respectively. Both approaches are generalizable for varying user scales. Moreover, the former can adapt to varying system utilities and antenna configurations by re-configuring the input and output module of the BERT model, while the latter, termed UBERT, can directly generalize to diverse tasks, due to a finer-grained tokenization strategy. Extensive simulation results demonstrate that the two proposed approaches can achieve near-optimal performance and outperform existing AI models across various beamforming optimization tasks, showcasing strong adaptability and generalizability.

eess.SY↗

arXiv2026

An Efficient, Reliable and Observable Collective Communication Library in Large-scale GPU Training Clusters

Large-scale LLM training requires collective communication libraries to exchange data among distributed GPUs. As a company dedicated to building and operating large-scale GPU training clusters, we encounter several practical limitations of NCCL in production, including 1) SM competition between computation and communication, 2) expensive restart costs under link failures, and 3) insufficient observability of transient collective communication anomalies. To address these challenges, we propose VCCL, an efficient, reliable, and observable collective communication library in large-scale GPU training clusters. VCCL removes SM-consuming P2P kernels by moving intra-node data movement and stream dependency enforcement to CPU threads and GPU copy engines. VCCL also introduces a primary-backup QP mechanism to tolerate frequent NIC port failures, and designs a window-based monitor to observe network anomalies at O(μs) level. We opensource VCCL and deploy it in production training clusters for several months. Compared with NCCL, VCCL improves training throughput by up to 5.28% and reduces massive GPU resource wastage through runtime fault tolerance and finegrained monitor. We also share experience and lessons we learned during the deployment of VCCL in large-scale clusters.

cs.DC↗

arXiv2026

NAPPure: Adversarial Purification for Robust Image Classification under Non-Additive Perturbations

Adversarial purification has achieved great success in combating adversarial image perturbations, which are usually assumed to be additive. However, non-additive adversarial perturbations such as blur, occlusion, and distortion are also common in the real world. Under such perturbations, existing adversarial purification methods are much less effective since they are designed to fit the additive nature. In this paper, we propose an extended adversarial purification framework named NAPPure, which can further handle non-additive perturbations. Specifically, we first establish the generation process of an adversarial image, and then disentangle the underlying clean image and perturbation parameters through likelihood maximization. Experiments on GTSRB and CIFAR-10 datasets show that NAPPure significantly boosts the robustness of image classification models against non-additive perturbations.

cs.CV↗

arXiv2026

Unsupervised Defect Detection for Surgical Instruments

Ensuring the safety of surgical instruments requires reliable detection of visual defects. However, manual inspection is prone to error, and existing automated defect detection methods, typically trained on natural/industrial images, fail to transfer effectively to the surgical domain. We demonstrate that simply applying or fine-tuning these approaches leads to issues: false positive detections arising from textured backgrounds, poor sensitivity to small, subtle defects, and inadequate capture of instrument-specific features due to domain shift. To address these challenges, we propose a versatile method that adapts unsupervised defect detection methods specifically for surgical instruments. By integrating background masking, a patch-based analysis strategy, and efficient domain adaptation, our method overcomes these limitations, enabling the reliable detection of fine-grained defects in surgical instrument imagery.

cs.CV↗

arXiv2026

Fault-Tolerant Quantum Key Distribution: Enabling Overclocked Modulation

Implementation security, higher generation rate, and lower cost are primary missions in the domain of quantum key distributions in recent years. However, simultaneously achieving robust security, high speed, and low cost often resembles an ``impossible triangle''. This is largely because the modulation system imposes a strict bandwidth limitation. Pushing a low-cost modulator to a high repetition frequency inevitably introduces correlations and misalignment, which can create security loopholes. Conversely, operating at a conservative rate fails to exploit the system's potential, while adopting ultra-high-bandwidth components is often expensive for practical implementation, forcing a perpetual trade-off among implementation security, key rate, and cost. In this work, we propose a comprehensive countermeasure to overcome this modulation bandwidth bottleneck. We present a protocol specifically designed to address the security loopholes arising from modulation imperfections, ensuring security even in overclocked modulation systems. Furthermore, we develop two practical techniques to characterize and mitigate the detrimental correlations. Our experimental setup demonstrates that the proposed method achieves the lowest correlated deviation reported in similar studies, while maintaining a high secret key rate using a bandwidth-limited modulation system. By simultaneously enhancing security, performance, and practicality, this work releases QKD systems from the traditional performance-cost trade-off in the near term, paving the way for widespread deployment. In the long run, this work can be readily integrated with high-bandwidth components to further push the boundaries of system performance.

quant-ph↗

arXiv2026

Extra-Merge: Tracing the Rank-1 Subspace of Model Merging in Language Model Pre-Training

Model merging has emerged as a lightweight paradigm for enhancing Large Language Models (LLMs), yet its underlying mechanisms remain poorly understood. In this work, we analyze late-stage pre-training trajectories and uncover a \textbf{Rank-1 Subspace} phenomenon: while raw optimization steps oscillate violently, consecutive \emph{merged} checkpoints collapse onto a stable, approximately one-dimensional linear manifold. We theoretically ground this observation in a \emph{river-valley} landscape analysis: averaging acts as a geometric low-pass filter that dampens high-curvature noise to reveal the optimal descent direction. Capitalizing on this insight, we propose \textbf{Extra-Merge}, a training-free strategy that extrapolates along this subspace to minimize loss without additional gradient updates. Extensive experiments across GPT-2 and LLaMA families (124M to 2B) demonstrate that Extra-Merge consistently outperforms standard merging baselines. Notably, it yields consistent zero-shot accuracy gains on Pythia-12B downstream tasks and generalizes effectively to the Muon optimizer \citep{jordan2024muon}.

cs.LG↗

arXiv2026

The Stability of Singular Distribution: A Spectral Perspective on the Two-Phase Dynamics of Language Model Pre-training

Large language model pre-training typically exhibits a two-phase trajectory: a fast initial loss drop followed by a prolonged slow improvement. We identify an underlying spectral phenomenon, Stability of Singular Distribution (SoSD), where the trace-normalized singular value spectrum stabilizes early, even as parameter matrices continue to evolve. We demonstrate that synchronization between SoSD and the slow-descent regime is widely observed across diverse architectures (GPT-2, LLaMA) and settings, including various schedules (Step-wise, WSD, Cosine Decay), weight decays, and optimizers (AdamW, Muon). By analyzing a simplified Transformer, we prove that growing weight norms inevitably precipitate an early SoSD threshold, after which the rate of loss decrease becomes theoretically bounded by the variation in the singular distribution. We further interpret strategies like WSD and Muon through their ability to modulate the SoSD scale, offering a spectral lens for understanding efficient pre-training dynamics.

cs.LG↗

arXiv2026

From Prompt Optimization to Multi-Dimensional Credibility Evaluation: Enhancing Trustworthiness of Chinese LLM-Generated Liver MRI Reports -- with Preliminary Extension to Lung Cancer

Large language models (LLMs) have demonstrated promising performance in generating diagnostic conclusions from imaging findings, thereby supporting radiology reporting, trainee education, and quality control. However, systematic guidance on how to optimize prompt design across different clinical contexts remains underexplored. Moreover, a comprehensive and standardized framework for assessing the trustworthiness of LLM-generated radiology reports is yet to be established. This study aims to enhance the trustworthiness of LLM-generated liver MRI reports by introducing a Multi-Dimensional Credibility Assessment (MDCA) framework and providing guidance on institution-specific prompt optimization. The proposed framework is applied to evaluate and compare the performance of several advanced LLMs, including Kimi-K2-Instruct-0905, Qwen3-235B-A22B-Instruct-2507, DeepSeek-V3, and ByteDance-Seed-OSS-36B-Instruct, using the SiliconFlow platform.

cs.AI↗

arXiv2026

X-Edit: Exact, Explicit, and Explainable Null-Space Editing for Medical Vision Transformers

Pre-trained Vision Transformers (ViTs) are increasingly deployed for medical image classification. However, correcting their inevitable failure cases in dynamic clinical scenarios poses a critical challenge. Conventional fine-tuning approaches inherently suffer from catastrophic forgetting, severely degrading previously acquired diagnostic capabilities. Such instability fundamentally compromises clinical safety. Addressing this vulnerability requires an active, controllable, and reliable intervention mechanism that is both theoretically grounded and inherently interpretable. To this end, we propose X-Edit (eXact, eXplicit, and eXplainable Editing), an efficient null-space model editing framework. X-Edit transitions the editing process from iterative gradient-based optimization to a theoretically grounded, closed-form solution. Specifically, we first explicitly localize the influential layers via causal tracing governing the erroneous prediction. Subsequently, we construct an orthogonal null-space projection matrix from a curated anchor set. By geometrically constraining the exact parameter update strictly within this null space, we provide mathematical guarantees that the intervention rectifies targeted errors without perturbing established diagnostic representations. Extensive evaluations on six medical imaging benchmarks demonstrate that X-Edit comprehensively suppresses catastrophic forgetting while achieving superior edit success rates. Our code is available at https://github.com/HenryLau7/X-Edit.

cs.CV↗

arXiv2026

InvariantCloud: A Globally Invariant, Uniquely Indexed Point Cloud Framework for Robust 6-DoF Tactile Pose Tracking

Recent advances in imitation learning and vision-language models highlight the need for high-fidelity tactile perception, with 6-DoF tactile object pose estimation providing a crucial foundation for precise robotic manipulation. We introduce InvariantCloud, a 6-DoF pose estimation framework that leverages the global invariance of surface marker constellations on vision-based tactile sensors. In contrast to recent approaches, our one-shot globally invariant point cloud registration suppresses cumulative drift and overcomes long-standing limitations in accurately estimating yaw (Z-axis) rotation. Experimental verifications show that InvariantCloud achieves superior yaw tracking accuracy and re-localization repeatability compared to existing benchmarks, demonstrating its precision and robustness in long-sequence manipulation tasks.

cs.RO↗