arXiv subjects

Feng Zhou

Publications and source records attributed to Feng Zhou.

At least 37 records · Page 2Linked to original sources

arXiv2026

Negative Binomial Variational Autoencoders for Overdispersed Latent Modeling

Although artificial neural networks are often described as brain-inspired, their representations typically rely on continuous activations, such as the continuous latent variables in variational autoencoders (VAEs), which limits their biological plausibility compared to the discrete spike-based signaling in real neurons. Extensions like the Poisson VAE introduce discrete count-based latents, but their equal mean-variance assumption fails to capture overdispersion in neural spikes, leading to less expressive and informative representations. To address this, we propose NegBio-VAE, a negative-binomial latent-variable model with a dispersion parameter for flexible spike count modeling. NegBio-VAE preserves interpretability while improving representation quality and training feasibility via novel KL estimation and reparameterization. Experiments on four datasets demonstrate that NegBio-VAE consistently achieves superior reconstruction and generation performance compared to competing single-layer VAE baselines, and yields robust, informative latent representations for downstream tasks. Extensive ablation studies are performed to verify the model's robustness w.r.t. various components. Our code is available at https://github.com/co234/NegBio-VAE.

cs.LG↗

arXiv2026

Magnetoelastic instabilities in kagome antiferromagnet Mn3-xGa

We present a systematic study of the structural, magnetic, and transport properties of hexagonal Mn3-xGa alloys, revealing a series of composition-controlled emergent phenomena. By tuning the Mn concentration, we uncover distinct lattice responses, including a zero thermal expansion-like volume compensation behavior in Mn-poor compositions and a magnetoelastic-driven, field-assisted structural phase transition in Mn-rich samples. These lattice instabilities are accompanied by correlated magnetic and transport anomalies, including metamagnetic transitions, negative magnetoresistance, and anomalous Hall sign reversal. First-principles calculations demonstrate that the Hall sign reversal originates from crystal-symmetry breaking rather than magnetic reorientation alone. Our results establish composition as the key control parameter governing magnetoelastic coupling in Mn3-xGa, providing a unified framework to tailor structural, magnetic, and topological transport properties in kagome antiferromagnets and reconcile previously disparate experimental observations.

cond-mat.mtrl-sci↗

arXiv2026

SpecTr-GBV: Multi-Draft Block Verification Accelerating Speculative Decoding

Autoregressive language models suffer from high inference latency due to their sequential decoding nature. Speculative decoding (SD) mitigates this by employing a lightweight draft model to propose candidate tokens, which are selectively verified by a larger target model. While existing methods either adopt multi-draft strategies to increase acceptance rates or block verification techniques to jointly verify multiple tokens, they remain limited by treating these improvements in isolation. In this work, we propose SpecTr-GBV, a novel SD method that unifies multi-draft and greedy block verification (GBV) into a single framework. By formulating the verification step as an optimal transport problem over draft and target token blocks, SpecTr-GBV improves both theoretical efficiency and empirical performance. We theoretically prove that SpecTr-GBV achieves the optimal expected acceptance length physically attainable within the framework of i.i.d. draft generation, and this bound improves as the number of drafts increases. Empirically, we evaluate SpecTr-GBV across five datasets and four baselines. Our method achieves superior speedup and significantly higher block efficiency while preserving output quality. In addition, we perform comprehensive ablation studies to evaluate the impact of various hyperparameters in the model.

cs.CL↗

arXiv2026

Byte-token Enhanced Language Models for Temporal Point Processes Analysis

Temporal Point Processes (TPPs) have been widely used for modeling event sequences on the Web, such as user reviews, social media posts, and online transactions. However, traditional TPP models often struggle to effectively incorporate the rich textual descriptions that accompany these events, while Large Language Models (LLMs), despite their remarkable text processing capabilities, lack mechanisms for handling the temporal dynamics inherent in Web-based event sequences. To bridge this gap, we introduce Language-TPP, a unified framework that seamlessly integrates TPPs with LLMs for enhanced Web event sequence modeling. Our key innovation is a novel temporal encoding mechanism that converts continuous time intervals into specialized byte-tokens, enabling direct integration with standard language model architectures for TPP modeling without requiring TPP-specific modifications. This approach allows Language-TPP to achieve state-of-the-art performance across multiple TPP benchmarks, including event time prediction and type prediction, on real-world Web datasets spanning e-commerce reviews, social media and online Q&A platforms. More importantly, we demonstrate that our unified framework unlocks new capabilities for TPP research: incorporating temporal information improves the quality of generated event descriptions, as evidenced by enhanced ROUGE-L scores, and better aligned sentiment distributions. Through comprehensive experiments, including qualitative analysis of learned distributions and scalability evaluations on long sequences, we show that Language-TPP effectively captures both temporal dynamics and textual patterns in Web user behavior, with important implications for content generation, user behavior understanding, and Web platform applications. Code is available at https://github.com/qykong/Language-TPP.

cs.CL↗

arXiv2026

Advanced Control of Electron Beams: Tailoring X-ray Production with Programmable Laser Shaping

Leveraging the full scientific capabilities of next-generation high-repetition-rate free-electron lasers requires programmable control over electron-beam properties at their source. The photoinjector drive laser defines the electron beam's initial six-dimensional phase-space distribution, yet has historically been limited to Gaussian or static flat-top profiles, with most manipulation occurring downstream. Here we demonstrate software-programmable ultraviolet pulse shaping at the LCLS-II photoinjector as a source-level actuator that complements traditional accelerator controls. Using a coupled architecture combining dispersion-controlled nonlinear frequency conversion with spatial-light-modulator spectral shaping, we generate user-defined temporal structures and observe their imprint on electron bunches through high-resolution time-domain diagnostics. Laser-imposed multi-peaked modulation persists through acceleration, magnetic compression, and undulator transport with shot-to-shot repeatability, producing clearly resolved current structure in the compressed beam. Variance-based reconstruction from transverse deflecting cavity measurements reveals structured X-ray emission profiles exhibiting temporal features consistent with the programmed laser waveform. By providing rapid, software-controlled reconfiguration of electron-beam initial conditions, this source-level control approach establishes a programmable upstream actuator for future adaptive optimization and autonomous facility operation at high-repetition-rate light sources.

physics.acc-ph↗

arXiv2026

On the Expressive Power of Transformers for Maxout Networks and Continuous Piecewise Linear Functions

Transformer networks have achieved remarkable empirical success across a wide range of applications, yet their theoretical expressive power remains insufficiently understood. In this paper, we study the expressive capabilities of Transformer architectures. We first establish an explicit approximation of maxout networks by Transformer networks while preserving comparable model complexity. As a consequence, Transformers inherit the universal approximation capability of ReLU networks under similar complexity constraints. Building on this connection, we develop a framework to analyze the approximation of continuous piecewise linear functions by Transformers and quantitatively characterize their expressivity via the number of linear regions, which grows exponentially with depth. Our analysis establishes a theoretical bridge between approximation theory for standard feedforward neural networks and Transformer architectures. It also yields structural insights into Transformers: self-attention layers implement max-type operations, while feedforward layers realize token-wise affine transformations.

cs.LG↗

arXiv2026

Long-range Modeling and Processing of Multimodal Event Sequences

Temporal point processes (TPPs) have emerged as powerful tools for modeling asynchronous event sequences. While recent advances have extended TPPs to handle textual information, existing approaches are limited in their ability to generate rich, multimodal content and reason about event dynamics. A key challenge is that incorporating multimodal data dramatically increases sequence length, hindering the ability of attention-based models to generate coherent, long-form textual descriptions that require long-range understanding. In this paper, we propose a novel framework that extends LLM-based TPPs to the visual modality, positioning text generation as a core capability alongside time and type prediction. Our approach addresses the long-context problem through an adaptive sequence compression mechanism based on temporal similarity, which reduces sequence length while preserving essential patterns. We employ a two-stage paradigm of pre-training on compressed sequences followed by supervised fine-tuning for downstream tasks. Extensive experiments, including on the challenging DanmakuTPP-QA benchmark, demonstrate that our method outperforms state-of-the-art baselines in both predictive accuracy and the quality of its generated textual analyses.

cs.CL↗

arXiv2026

A Tilted Seesaw: Revisiting Autoencoder Trade-off for Controllable Diffusion

In latent diffusion models, the autoencoder (AE) is typically expected to balance two capabilities: faithful reconstruction and a generation-friendly latent space (e.g., low gFID). In recent ImageNet-scale AE studies, we observe a systematic bias toward generative metrics in handling this trade-off: reconstruction metrics are increasingly under-reported, and ablation-based AE selection often favors the best-gFID configuration even when reconstruction fidelity degrades. We theoretically analyze why this gFID-dominant preference can appear unproblematic for ImageNet generation, yet becomes risky when scaling to controllable diffusion: AEs can induce condition drift, which limits achievable condition alignment. Meanwhile, we find that reconstruction fidelity, especially instance-level measures, better indicates controllability. We empirically validate the impact of tilted autoencoder evaluation on controllability by studying several recent ImageNet AEs. Using a multi-dimensional condition-drift evaluation protocol reflecting controllable generation tasks, we find that gFID is only weakly predictive of condition preservation, whereas reconstruction-oriented metrics are substantially more aligned. ControlNet experiments further confirm that controllability tracks condition preservation rather than gFID. Overall, our results expose a gap between ImageNet-centric AE evaluation and the requirements of scalable controllable diffusion, offering practical guidance for more reliable benchmarking and model selection.

cs.CV↗

arXiv2026

Mud-Standoff Effect Correction Based on Open-Short Calibration and Resistivity Consistency-Constrained Iterative Inversion for Oil-Based Mud Imagers

We propose a mud-standoff effect correction method and a set of approximate apparent resistivity inversion methods suitable for oil-based mud micro-resistivity imaging logging. To calibrate the influence of the mud layer on electrode measurement signals, this study integrates the Open-Short calibration(OSC) method with the three-layer impedance model of the oil-based mud resistivity imager. By treating the electrode and the mud layer as an integrated whole and simulating the open/short-circuit states via the finite element method, the independent extraction of the mud layer impedance signal is achieved, and the formation impedance signal is separated from the total impedance. For fast inversion of formation resistivity, standoff thickness (mud layer thickness), and relative permittivity of formation, a resistivity consistency-constrained iterative inversion method is further proposed. In this method, the formation impedance is first converted into resistivity, and then the consistency residual of the resistivity at different frequencies is used as the objective function to invert the approximate apparent resistivity of the formation through an iterative optimization algorithm. The effectiveness of the finite element-simulated OSC method, the objective function construction scheme, and the consistency inversion method is verified through both numerical models and an example of field data.

physics.geo-ph↗

arXiv2026

Upstream Laser-based Longitudinal Enhancement of Relativistic Photoelectrons

Controlling the longitudinal phase space of high-brightness relativistic electron beams is crucial for advancing a broad spectrum of charged-particle-based instrumentation and scientific frontiers. A generalized method for achieving this control involves manipulating the photoemission laser's temporal distribution at the picosecond level, a long-standing technical challenge. Recent developments in laser shaping have enabled the creation of high-power, picosecond-scale symmetrical and asymmetrical temporal profiles, capable of fine-tuning complex space-charge dynamics and external field effects in relativistic charged-particle beams. Here, we demonstrate that rather than deviations from theorized, idealized laser distributions, a controlled asymmetry can be harnessed to counteract accelerator-induced distortions. By implementing spatiotemporal shaping of the ultraviolet photocathode laser at the LCLS-II superconducting injector, we achieve deterministic control over the longitudinal phase space without downstream corrections. We find that this optical asymmetry induces a self-linearizing effect across both low (40 pC) and high (80 pC) charge regimes, effectively suppressing nonlinear compression and energy chirp. Consequently, this approach is expected to preserve a low emittance comparable to that of ideal flattop or regular Gaussian profiles, while delivering superior current uniformity and shot-to-shot stability. These results establish spatiotemporal laser shaping as a compact, generalizable tool for directly optimizing beam brightness at the source.

physics.optics↗

arXiv2026

Controllable Generation with Text-to-Image Diffusion Models: A Survey

In the rapidly advancing realm of visual generation, diffusion models have revolutionized the landscape, marking a significant shift in capabilities with their impressive text-guided generative functions. However, relying solely on text for conditioning these models does not fully cater to the varied and complex requirements of different applications and scenarios. Acknowledging this shortfall, a variety of studies aim to control pre-trained text-to-image (T2I) models to support novel conditions. In this survey, we undertake a thorough review of the literature on controllable generation with T2I diffusion models, covering both the theoretical foundations and practical advancements in this domain. Our review begins with a brief introduction to the basics of denoising diffusion probabilistic models (DDPMs) and widely used T2I diffusion models. We then reveal the controlling mechanisms of diffusion models, theoretically analyzing how novel conditions are introduced into the denoising process for conditional generation. Additionally, we offer a detailed overview of research in this area, organizing it into distinct categories from the condition perspective: generation with specific conditions, generation with multiple conditions, and universal controllable generation. For an exhaustive list of the controllable generation literature surveyed, please refer to our curated repository at https://github.com/PRIV-Creation/Awesome-Controllable-T2I-Diffusion-Models.

cs.CV↗

arXiv2025

Score Matching for Estimating Finite Point Processes

Score matching estimators have garnered significant attention in recent years because they eliminate the need to compute normalizing constants, thereby mitigating the computational challenges associated with maximum likelihood estimation (MLE).While several studies have proposed score matching estimators for point processes, this work highlights the limitations of these existing methods, which stem primarily from the lack of a mathematically rigorous analysis of how score matching behaves on finite point processes -- special random configurations on bounded spaces where many of the usual assumptions and properties of score matching no longer hold. To this end, we develop a formal framework for score matching on finite point processes via Janossy measures and, within this framework, introduce an (autoregressive) weighted score-matching estimator, whose statistical properties we analyze in classical parametric settings. For general nonparametric (e.g., deep) point process models, we show that score matching alone does not uniquely identify the ground-truth distribution due to subtle normalization issues, and we propose a simple survival-classification augmentation that yields a complete, integration-free training objective for any intensity-based point process model for spatio-temporal case. Experiments on synthetic and real-world temporal and spatio-temporal datasets, demonstrate that our method accurately recovers intensities and achieves performance comparable to MLE with better efficiency.

cs.LG↗

arXiv2025

ResDiT: Evoking the Intrinsic Resolution Scalability in Diffusion Transformers

Leveraging pre-trained Diffusion Transformers (DiTs) for high-resolution (HR) image synthesis often leads to spatial layout collapse and degraded texture fidelity. Prior work mitigates these issues with complex pipelines that first perform a base-resolution (i.e., training-resolution) denoising process to guide HR generation. We instead explore the intrinsic generative mechanisms of DiTs and propose ResDiT, a training-free method that scales resolution efficiently. We identify the core factor governing spatial layout, position embeddings (PEs), and show that the original PEs encode incorrect positional information when extrapolated to HR, which triggers layout collapse. To address this, we introduce a PE scaling technique that rectifies positional encoding under resolution changes. To further remedy low-fidelity details, we develop a local-enhancement mechanism grounded in base-resolution local attention. We design a patch-level fusion module that aggregates global and local cues, together with a Gaussian-weighted splicing strategy that eliminates grid artifacts. Comprehensive evaluations demonstrate that ResDiT consistently delivers high-fidelity, high-resolution image synthesis and integrates seamlessly with downstream tasks, including spatially controlled generation.

cs.CV↗

arXiv2025

Fair Bayesian Data Selection via Generalized Discrepancy Measures

Fairness concerns are increasingly critical as machine learning models are deployed in high-stakes applications. While existing fairness-aware methods typically intervene at the model level, they often suffer from high computational costs, limited scalability, and poor generalization. To address these challenges, we propose a Bayesian data selection framework that ensures fairness by aligning group-specific posterior distributions of model parameters and sample weights with a shared central distribution. Our framework supports flexible alignment via various distributional discrepancy measures, including Wasserstein distance, maximum mean discrepancy, and $f$-divergence, allowing geometry-aware control without imposing explicit fairness constraints. This data-centric approach mitigates group-specific biases in training data and improves fairness in downstream tasks, with theoretical guarantees. Experiments on benchmark datasets show that our method consistently outperforms existing data selection and model-based fairness methods in both fairness and accuracy.

cs.LG↗

arXiv2025

Beam Halo Formation via Longitudinal-Transverse Coupling in Continuous-Wave Photoinjectors

Beam halo formation poses a critical challenge for high-repetition-rate continuous-wave (CW) free-electron lasers (FELs), directly affecting beam quality and machine protection, as observed during the LCLS-II commissioning. We identify and experimentally validate a previously unrecognized three-step mechanism for halo generation in the photoinjector, arising from coupled longitudinal-transverse dynamics in the low-energy beam. Theoretical analysis reveals that (i) the RF buncher induces an energy-radius correlation, (ii) velocity bunching transforms this correlation into hollowed density structures in the bunch head and tail, and (iii) differential overfocusing of these hollowed regions by downstream focusing forms the observed halo. This mechanism is confirmed by particle-in-cell simulations and direct experimental measurements, including controlled formation of a core-ring profile via solenoid tuning. The results establish the physical origin of the halo and demonstrate a mitigation via buncher compression tuning that reduces halo and downstream loss, supporting sustained high-repetition-rate FEL operation.

physics.acc-ph↗

arXiv2025

DanmakuTPPBench: A Multi-modal Benchmark for Temporal Point Process Modeling and Understanding

We introduce DanmakuTPPBench, a comprehensive benchmark designed to advance multi-modal Temporal Point Process (TPP) modeling in the era of Large Language Models (LLMs). While TPPs have been widely studied for modeling temporal event sequences, existing datasets are predominantly unimodal, hindering progress in models that require joint reasoning over temporal, textual, and visual information. To address this gap, DanmakuTPPBench comprises two complementary components: (1) DanmakuTPP-Events, a novel dataset derived from the Bilibili video platform, where user-generated bullet comments (Danmaku) naturally form multi-modal events annotated with precise timestamps, rich textual content, and corresponding video frames; (2) DanmakuTPP-QA, a challenging question-answering dataset constructed via a novel multi-agent pipeline powered by state-of-the-art LLMs and multi-modal LLMs (MLLMs), targeting complex temporal-textual-visual reasoning. We conduct extensive evaluations using both classical TPP models and recent MLLMs, revealing significant performance gaps and limitations in current methods' ability to model multi-modal event dynamics. Our benchmark establishes strong baselines and calls for further integration of TPP modeling into the multi-modal language modeling landscape. Project page: https://github.com/FRENKIE-CHIANG/DanmakuTPPBench

cs.CL↗

arXiv2025

TPP-SD: Accelerating Transformer Point Process Sampling with Speculative Decoding

We propose TPP-SD, a novel approach that accelerates Transformer temporal point process (TPP) sampling by adapting speculative decoding (SD) techniques from language models. By identifying the structural similarities between thinning algorithms for TPPs and speculative decoding for language models, we develop an efficient sampling framework that leverages a smaller draft model to generate multiple candidate events, which are then verified by the larger target model in parallel. TPP-SD maintains the same output distribution as autoregressive sampling while achieving significant acceleration. Experiments on both synthetic and real datasets demonstrate that our approach produces samples from identical distributions as standard methods, but with 2-6$\times$ speedup. Our ablation studies analyze the impact of hyperparameters such as draft length and draft model size on sampling efficiency. TPP-SD bridges the gap between powerful Transformer TPP models and the practical need for rapid sequence sampling.

cs.LG↗

arXiv2025

Initialize to Generalize: A Stronger Initialization Pipeline for Sparse-View 3DGS

Sparse-view 3D Gaussian Splatting (3DGS) often overfits to the training views, leading to artifacts like blurring in novel view rendering. Prior work addresses it either by enhancing the initialization (\emph{i.e.}, the point cloud from Structure-from-Motion (SfM)) or by adding training-time constraints (regularization) to the 3DGS optimization. Yet our controlled ablations reveal that initialization is the decisive factor: it determines the attainable performance band in sparse-view 3DGS, while training-time constraints yield only modest within-band improvements at extra cost. Given initialization's primacy, we focus our design there. Although SfM performs poorly under sparse views due to its reliance on feature matching, it still provides reliable seed points. Thus, building on SfM, our effort aims to supplement the regions it fails to cover as comprehensively as possible. Specifically, we design: (i) frequency-aware SfM that improves low-texture coverage via low-frequency view augmentation and relaxed multi-view correspondences; (ii) 3DGS self-initialization that lifts photometric supervision into additional points, compensating SfM-sparse regions with learned Gaussian centers; and (iii) point-cloud regularization that enforces multi-view consistency and uniform spatial coverage through simple geometric/visibility priors, yielding a clean and reliable point cloud. Our experiments on LLFF and Mip-NeRF360 demonstrate consistent gains in sparse-view settings, establishing our approach as a stronger initialization strategy. Code is available at https://github.com/zss171999645/ItG-GS.

cs.CV↗