Searcharxiv⌕ Search

arXiv subjects

Yifan Chen

Publications and source records attributed to Yifan Chen.

At least 73 records · Page 4Linked to original sources
arXiv2026

Stellar Superradiance and Low-Energy Absorption in Dense Nuclear Media

Ultralight bosons such as axions and dark photons are well-motivated hypothetical particles, whose couplings to ordinary matter can be effectively constrained by stellar cooling. Limits on these interactions can be obtained by demanding that their emission from the stellar interior does not lead to excessive energy loss. An intriguing question is whether the same microphysical couplings can also be probed through neutron star superradiance, in which gravitationally bound bosonic modes grow exponentially by extracting rotational energy from the star. Although both processes originate from boson-matter interactions, they probe very different kinematic regimes. Stellar cooling probes boson emission at thermal wavelengths, while superradiance is governed by modes whose wavelength is comparable to or greater than the size of the star. Previous work has attempted to relate the microphysical neutron-nucleon scattering and inverse-bremsstrahlung absorption rates directly to the macroscopic growth rate of superradiant bound states. In this work, we re-examine this connection and show that a naive extrapolation of the microphysical absorption rate to the superradiant regime would imply superradiant rates comparable to astrophysical timescales characterised by pulsar spindown. These naive rates are especially high for vector fields. However, we demonstrate that this conclusion changes once collective multiple-scattering effects in dense nuclear matter are taken into account. Repeated nucleon collisions modify the effective low-energy absorption experienced by the bosonic bound state, strongly suppressing the rate relevant for superradiance.

hep-ph↗
arXiv2026

Asteroseismic Diagnostics for Red Giants with Kepler: Measuring epsilon and Small Frequency Separations in 16,000 Stars

Asteroseismic studies of red giants have primarily relied on two global parameters: the large frequency separation (Dnu) and the frequency of maximum power (numax). Meanwhile, the p-mode phase shift (epsilon) and small frequency separations (dnu01, dnu02), which offer additional constraints on stellar interiors, remain underexplored due to measurement challenges. Here we develop an automated pipeline based on collapsed echelle diagrams and apply it to about 16,000 Kepler red giants, jointly measuring Dnu, epsilon, dnu01, and dnu02 and assembling the largest homogeneous catalogue of these quantities to date, together with updated Dnu values and formal internal uncertainties. Using this catalogue, we quantify evolutionary trends across the red-giant branch and core-helium-burning phase. We find that dnu02/Dnu stays nearly constant for RGB stars and, for core-helium-burning stars, organises into two sequences that are systematically offset but partially overlap, broadly separating stars in the red-clump and secondary-clump regimes. We also trace the mass- and metallicity-dependent helium-flash transition. Meanwhile, epsilon follows a single Dnu-epsilon relation common to both evolutionary phases. Comparisons with stellar-evolution models reveal systematic offsets in epsilon and dnu01, which we interpret as signatures of near-surface and outer-envelope modelling deficiencies. These comparisons further suggest that dipole-mode small separations are sensitive to mode-dependent surface terms in evolved stars. Overall, our results demonstrate that epsilon and the small separations provide important diagnostics of core structure, convective-boundary mixing, and helium ignition that are complementary to those provided by Dnu and numax alone. The resulting catalogue offers a reference for testing and calibrating future stellar-evolution models.

astro-ph.SR↗
arXiv2026

SynLeaF: A Dual-Stage Multimodal Fusion Framework for Synthetic Lethality Prediction Across Pan- and Single-Cancer Contexts

Accurate prediction of synthetic lethality (SL) is important for guiding the development of cancer drugs and therapies. SL prediction faces significant challenges in the effective fusion of heterogeneous multi-source data. Existing multimodal methods often suffer from "modality laziness" due to disparate convergence speeds, which hinders the exploitation of complementary information. This is also one reason why most existing SL prediction models cannot perform well on both pan-cancer and single-cancer SL pair prediction. In this study, we propose SynLeaF, a dual-stage multimodal fusion framework for SL prediction across pan- and single-cancer contexts. The framework employs a VAE-based cross-encoder with a product of experts mechanism to fuse four omics data types (gene expression, mutation, methylation, and CNV), while simultaneously utilizing a relational graph convolutional network to capture structured gene representations from biomedical knowledge graphs. To mitigate modality laziness, SynLeaF introduces a dual-stage training mechanism employing featurelevel knowledge distillation with adaptive uni-modal teacher and ensemble strategies. In extensive experiments across eight specific cancer types and a pancancer dataset, SynLeaF achieves superior performance in 17 out of 19 scenarios. Ablation studies and gradient analyses further validate the critical contributions of the proposed fusion and distillation mechanisms to model robustness and generalization. To facilitate community use, a web server is available at https://synleaf.bioinformatics-lilab.cn.

q-bio.GN↗
arXiv2026

Advancing Visual Reliability: Color-Accurate Underwater Image Enhancement for Real-Time Underwater Missions

Underwater image enhancement plays a crucial role in providing reliable visual information for underwater platforms, since strong absorption and scattering in water-related environments generally lead to image quality degradation. Existing high-performance methods often rely on complex architectures, which hinder deployment on underwater devices. Lightweight methods often sacrifice quality for speed and struggle to handle severely degraded underwater images. To address this limitation, we present a real-time underwater image enhancement framework with accurate color restoration. First, an Adaptive Weighted Channel Compensation module is introduced to achieve dynamic color recovery of the red and blue channels using the green channel as a reference anchor. Second, we design a Multi-branch Re-parameterized Dilated Convolution that employs multi-branch fusion during training and structural re-parameterization during inference, enabling large receptive field representation with low computational overhead. Finally, a Statistical Global Color Adjustment module is employed to optimize overall color performance based on statistical priors. Extensive experiments on eight datasets demonstrate that the proposed method achieves state-of-the-art performance across seven evaluation metrics. The model contains only 3,880 inference parameters and achieves an inference speed of 409 FPS. Our method improves the UCIQE score by 29.7% under diverse environmental conditions, and the deployment on ROV platforms and performance gains in downstream tasks further validate its superiority for real-time underwater missions.

cs.CV↗
arXiv2026

Fisher-Rao Gradient Flow: Geodesic Convexity and Functional Inequalities

The dynamics of probability density functions have been extensively studied in computational science and engineering to understand physical phenomena and facilitate algorithmic design. Of particular interest are dynamics formulated as gradient flows of energy functionals under the Wasserstein metric. The development of functional inequalities, such as the log-Sobolev inequality, plays a pivotal role in analyzing the convergence of these dynamics. This paper aims to extend the success of functional inequality techniques to dynamics that are gradient flows under the Fisher-Rao metric, with various $f$-divergences serving as energy functionals. Such dynamics take the form of nonlocal differential equations, for which existing analyses critically rely on explicit solution formulas in special cases. We provide a comprehensive study of functional inequalities and the relevant geodesic convexity for Fisher-Rao gradient flows under minimal assumptions. A notable feature of our functional inequalities is their independence from the log-concavity or log-Sobolev constants of the target distribution. Consequently, the convergence rate of the dynamics (assuming well-posedness) remains uniform across general target distributions.

math.AP↗
arXiv2026

Mitigating Long-Tail Bias in HOI Detection via Adaptive Diversity Cache

Human-Object Interaction (HOI) detection is a fundamental task in computer vision, empowering machines to comprehend human-object relationships in diverse real-world scenarios. Recent advances in VLMs have significantly improved HOI detection by leveraging rich cross-modal representations. However, most existing VLM-based approaches rely heavily on additional training or prompt tuning, resulting in substantial computational overhead and limited scalability, particularly in long-tailed scenarios where rare interactions are severely underrepresented. In this paper, we propose the Adaptive Diversity Cache (ADC) module, a novel training-free and plug-and-play mechanism designed to mitigate long-tail bias in HOI detection. ADC constructs class-specific caches that accumulate high-confidence and diverse feature representations during inference. The method incorporates adaptive capacity allocation favoring rare categories and dynamic feature augmentation to enable robust prediction calibration without requiring additional training or fine-tuning. Extensive experiments on HICO-DET and V-COCO datasets show that ADC consistently improves existing HOI detectors, particularly enhancing rare category detection while preserving overall performance. These findings confirm the effectiveness of ADC as a training-free, plug-and-play solution for long-tail bias mitigation.

cs.CV↗
arXiv2026

AdaptOVCD: Training-Free Open-Vocabulary Remote Sensing Change Detection via Adaptive Information Fusion

Remote sensing change detection plays a pivotal role in domains such as environmental monitoring, urban planning, and disaster assessment. However, existing methods typically rely on predefined categories and large-scale pixel-level annotations, which limit their generalization and applicability in open-world scenarios. To address these limitations, this paper proposes AdaptOVCD, a training-free Open-Vocabulary Change Detection (OVCD) architecture based on dual-dimensional multi-level information fusion. The framework integrates multi-level information fusion across data, feature, and decision levels vertically while incorporating targeted adaptive designs horizontally, achieving deep synergy among heterogeneous pre-trained models to effectively mitigate error propagation. Specifically, (1) at the data level, Adaptive Radiometric Alignment (ARA) fuses radiometric statistics with original texture features and synergizes with SAM-HQ to achieve radiometrically consistent segmentation; (2) at the feature level, Adaptive Change Thresholding (ACT) combines global difference distributions with edge structure priors and leverages DINOv3 to achieve robust change detection; (3) at the decision level, Adaptive Confidence Filtering (ACF) integrates semantic confidence with spatial constraints and collaborates with DGTRS-CLIP to achieve high-confidence semantic identification. Comprehensive evaluations across nine scenarios demonstrate that AdaptOVCD detects arbitrary category changes in a zero-shot manner, significantly outperforming existing training-free methods. Meanwhile, it achieves 84.89\% of the fully-supervised performance upper bound in cross-dataset evaluations and exhibits superior generalization capabilities. The code is available at https://github.com/Dmygithub/AdaptOVCD.

cs.CV↗
arXiv2026

Inference on inner galaxy structure via gravitational waves from supermassive binaries

The detection of a stochastic gravitational wave background by pulsar-timing arrays indicates the presence of a population of supermassive black hole binaries. Although the observed spectrum generally matches predictions for orbital evolution driven by gravitational-wave emission in circular orbits, there is a preference for a spectral turnover at the lowest observed frequencies, which may point to substantial hardening during a transition from early environmental influences to later stages dominated by emission. In the vicinity of these binaries, the ejection of stars or dark matter particles through gravitational three-body slingshots efficiently extracts orbital energy, leading to a low-frequency turnover in the spectrum. Here we model how the gravitational-wave spectrum depends on the initial inner galactic profile before scouring by binary ejections while accounting for a range of initial binary eccentricities. By analysing the NANOGrav 15-year data, we find that a parsec-scale galactic-centre density of around $10^6 M_{\odot} \mathrm{pc}^{-3}$ is favoured across most of the parameter space, thus shedding light on the environmental effects that shape black hole evolution and the combined matter density near galaxy centres.

astro-ph.HE↗
arXiv2026

LegalOne: A Family of Foundation Models for Reliable Legal Reasoning

While Large Language Models (LLMs) have demonstrated impressive general capabilities, their direct application in the legal domain is often hindered by a lack of precise domain knowledge and complexity of performing rigorous multi-step judicial reasoning. To address this gap, we present LegalOne, a family of foundational models specifically tailored for the Chinese legal domain. LegalOne is developed through a comprehensive three-phase pipeline designed to master legal reasoning. First, during mid-training phase, we propose Plasticity-Adjusted Sampling (PAS) to address the challenge of domain adaptation. This perplexity-based scheduler strikes a balance between the acquisition of new knowledge and the retention of original capabilities, effectively establishing a robust legal foundation. Second, during supervised fine-tuning, we employ Legal Agentic CoT Distillation (LEAD) to distill explicit reasoning from raw legal texts. Unlike naive distillation, LEAD utilizes an agentic workflow to convert complex judicial processes into structured reasoning trajectories, thereby enforcing factual grounding and logical rigor. Finally, we implement a Curriculum Reinforcement Learning (RL) strategy. Through a progressive reinforcement process spanning memorization, understanding, and reasoning, LegalOne evolves from simple pattern matching to autonomous and reliable legal reasoning. Experimental results demonstrate that LegalOne achieves state-of-the-art performance across a wide range of legal tasks, surpassing general-purpose LLMs with vastly larger parameter counts through enhanced knowledge density and efficiency. We publicly release the LegalOne weights and the LegalKit evaluation framework to advance the field of Legal AI, paving the way for deploying trustworthy and interpretable foundation models in high-stakes judicial applications.

cs.CL↗
arXiv2026

PLawBench: A Rubric-Based Benchmark for Evaluating LLMs in Real-World Legal Practice

As large language models (LLMs) are increasingly applied to legal domain-specific tasks, evaluating their ability to perform legal work in real-world settings has become essential. However, existing legal benchmarks rely on simplified and highly standardized tasks, failing to capture the ambiguity, complexity, and reasoning demands of real legal practice. Moreover, prior evaluations often adopt coarse, single-dimensional metrics and do not explicitly assess fine-grained legal reasoning. To address these limitations, we introduce PLawBench, a Practical Law Benchmark designed to evaluate LLMs in realistic legal practice scenarios. Grounded in real-world legal workflows, PLawBench models the core processes of legal practitioners through three task categories: public legal consultation, practical case analysis, and legal document generation. These tasks assess a model's ability to identify legal issues and key facts, perform structured legal reasoning, and generate legally coherent documents. PLawBench comprises 850 questions across 13 practical legal scenarios, with each question accompanied by expert-designed evaluation rubrics, resulting in approximately 12,500 rubric items for fine-grained assessment. Using an LLM-based evaluator aligned with human expert judgments, we evaluate 10 state-of-the-art LLMs. Experimental results show that none achieves strong performance on PLawBench, revealing substantial limitations in the fine-grained legal reasoning capabilities of current LLMs and highlighting important directions for future evaluation and development of legal LLMs. Data is available at: https://github.com/skylenage/PLawbench.

cs.CL↗
arXiv2026

Accumulation of Sub-Sampling Matrices with Applications to Statistical Computation

With appropriately chosen sampling probabilities, sampling-based random projection can be used to implement large-scale statistical methods, substantially reducing computational cost while maintaining low statistical error. However, computing optimal sampling probabilities is often itself expensive, and in practice one typically resorts to suboptimal schemes. This generally leads to increased time and space costs, as more subsamples are required and the resulting projection matrices become larger, thereby making the inference procedure more computationally demanding. In this paper, we extend the framework of sampling-based random projection and propose a new projection method, \emph{accumulative sub-sampling}. By carefully accumulating multiple such projections, accumulative sub-sampling improves statistical efficiency while controlling the effective matrix size throughout the statistical computation. On the theoretical side, we quantify how the quality of the subsampling scheme affects the error in approximating matrix products and positive semidefinite matrices, and show how the proposed accumulation strategy mitigates this effect. Moreover, we apply our method to statistical models involving intensive matrix operations, such as eigendecomposition in spectral clustering and matrix inversion in kernel ridge regression, and demonstrate that reducing the effective matrix size leads to substantial computational savings. Numerical experiments across a range of problems further show that our approach consistently improves computational efficiency compared to existing random projection baselines under suboptimal sampling schemes.

stat.ML↗
arXiv2025

New affine invariant ensemble samplers and their dimensional scaling

We introduce new affine invariant ensemble Markov chain Monte Carlo (MCMC) samplers that are easy to construct and improve upon existing methods, especially for high-dimensional problems. We first propose a simple derivative-free side move sampler that improves upon popular samplers in the \texttt{emcee} package by generating more effective proposal directions. We then develop a class of derivative-based affine invariant ensemble Hamiltonian Monte Carlo (HMC) samplers based on antisymmetric preconditioning using complementary ensembles, which outperform standard, non-affine-invariant HMC when sampling highly anisotropic distributions. We provide asymptotic scaling analysis for high-dimensional Gaussian targets to further elucidate the properties of these affine invariant ensemble samplers. In particular, with derivative information, the affine invariant ensemble HMC can scale much better with dimension compared to derivative-free ensemble samplers.

stat.CO↗
arXiv2025

Towards Robust Influence Functions with Flat Validation Minima

The Influence Function (IF) is a widely used technique for assessing the impact of individual training samples on model predictions. However, existing IF methods often fail to provide reliable influence estimates in deep neural networks, particularly when applied to noisy training data. This issue does not stem from inaccuracies in parameter change estimation, which has been the primary focus of prior research, but rather from deficiencies in loss change estimation, specifically due to the sharpness of validation risk. In this work, we establish a theoretical connection between influence estimation error, validation set risk, and its sharpness, underscoring the importance of flat validation minima for accurate influence estimation. Furthermore, we introduce a novel estimation form of Influence Function specifically designed for flat validation minima. Experimental results across various tasks validate the superiority of our approach.

cs.LG↗
arXiv2025

Optimizing LVLMs with On-Policy Data for Effective Hallucination Mitigation

Recently, large vision-language models (LVLMs) have risen to be a promising approach for multimodal tasks. However, principled hallucination mitigation remains a critical challenge.In this work, we first analyze the data generation process in LVLM hallucination mitigation and affirm that on-policy data significantly outperforms off-policy data, which thus calls for efficient and reliable preference annotation of on-policy data. We then point out that, existing annotation methods introduce additional hallucination in training samples, which may enhance the model's hallucination patterns, to address this problem, we propose training a hallucination classifier giving binary annotations, which guarantee clean chosen samples for the subsequent alignment. To further harness of the power of on-policy data, we design a robust iterative direct preference optimization (DPO) algorithm adopting a dynamic sample reweighting scheme. We conduct comprehensive experiments on three benchmarks with comparison to 8 state-of-the-art baselines. In particular, our approach reduces the hallucination rate of LLaVA-1.5-7B on MMHalBench by 50.8% and the average hallucination rate on Object HalBench by 79.5%; more significantly, our method fully taps into the potential of open-source models, enabling LLaVA-1.5-13B to even surpass the performance of GPT-4V.

cs.CV↗
arXiv2025

When Human Preferences Flip: An Instance-Dependent Robust Loss for RLHF

Quality of datasets plays an important role in large language model (LLM) alignment. In collecting human feedback, however, preference flipping is ubiquitous and causes corruption in data annotation; the issue necessitates the alignment algorithms with improved robustness against potential flipped pairs. To this end, this paper introduces a Flipping-Aware Direct Preference Optimization (FA-DPO) algorithm tailored to preference flipping from a reinforcement learning with human feedback (RLHF) perspective. We dissect the inherent human intention model and the preference flipping mechanism introduced by external factors as two distinct stages; in the latter, we introduce an instance-dependent flipping probability on the basis of the Bradley-Terry (BT) model. Further, by leveraging features relevant to preference annotation, we capture uncertainty in judgments and model preference flipping patterns. In practice, we design a simple yet efficient iterative optimization algorithm compatible with the original RLHF and DPO algorithms. In our experiments, we investigate the instance-dependent preference flipping model under multiple circumstances for evaluation of our proposed method, as well as other baseline methods.

cs.AI↗
arXiv2025

Heralded Emission Detection in InAs/ZnSe Quantum Dot Solids Using Time-Correlated Photons

Harnessing quantum correlations between photons is an emerging frontier in optical spectroscopy, yet experimental demonstrations have largely remained limited to molecular systems at room temperature. Here, we investigate heralded emission detection (HED) under continuous-wave entangled-photon excitation of near-infrared (NIR)-emitting colloidal III-V quantum dot (QD) solids at low temperatures. We demonstrate the advantages of superconducting nanowire single-photon detectors (SNSPDs) for high time resolution ($\sim$72 ps) and large-area NIR avalanche photodiodes (APDs) for high emission count rates ($\sim$2000 cps). Second-order photon-correlation analysis reveals exciton lifetimes and fine-structure energy splittings. These results establish NIR colloidal QDs as a bright, tunable model system for quantum-light spectroscopy and highlight their compatibility with optical cavities as a further experimental control parameter.

physics.optics↗
arXiv2025

Active Negative Loss: A Robust Framework for Learning with Noisy Labels

Deep supervised learning has achieved remarkable success across a wide range of tasks, yet it remains susceptible to overfitting when confronted with noisy labels. To address this issue, noise-robust loss functions offer an effective solution for enhancing learning in the presence of label noise. In this work, we systematically investigate the limitation of the recently proposed Active Passive Loss (APL), which employs Mean Absolute Error (MAE) as its passive loss function. Despite the robustness brought by MAE, one of its key drawbacks is that it pays equal attention to clean and noisy samples; this feature slows down convergence and potentially makes training difficult, particularly in large-scale datasets. To overcome these challenges, we introduce a novel loss function class, termed Normalized Negative Loss Functions (NNLFs), which serve as passive loss functions within the APL framework. NNLFs effectively address the limitations of MAE by concentrating more on memorized clean samples. By replacing MAE in APL with our proposed NNLFs, we enhance APL and present a new framework called Active Negative Loss (ANL). Moreover, in non-symmetric noise scenarios, we propose an entropy-based regularization technique to mitigate the vulnerability to the label imbalance. Extensive experiments demonstrate that the new loss functions adopted by our ANL framework can achieve better or comparable performance to state-of-the-art methods across various label noise types and in image segmentation tasks. The source code is available at: https://github.com/Virusdoll/Active-Negative-Loss.

cs.CV↗
arXiv2025

A Generative Data Framework with Authentic Supervision for Underwater Image Restoration and Enhancement

Underwater image restoration and enhancement are crucial for correcting color distortion and restoring image details, thereby establishing a fundamental basis for subsequent underwater visual tasks. However, current deep learning methodologies in this area are frequently constrained by the scarcity of high-quality paired datasets. Since it is difficult to obtain pristine reference labels in underwater scenes, existing benchmarks often rely on manually selected results from enhancement algorithms, providing debatable reference images that lack globally consistent color and authentic supervision. This limits the model's capabilities in color restoration, image enhancement, and generalization. To overcome this limitation, we propose using in-air natural images as unambiguous reference targets and translating them into underwater-degraded versions, thereby constructing synthetic datasets that provide authentic supervision signals for model learning. Specifically, we establish a generative data framework based on unpaired image-to-image translation, producing a large-scale dataset that covers 6 representative underwater degradation types. The framework constructs synthetic datasets with precise ground-truth labels, which facilitate the learning of an accurate mapping from degraded underwater images to their pristine scene appearances. Extensive quantitative and qualitative experiments across 6 representative network architectures and 3 independent test sets show that models trained on our synthetic data achieve comparable or superior color restoration and generalization performance to those trained on existing benchmarks. This research provides a reliable and scalable data-driven solution for underwater image restoration and enhancement. The generated dataset is publicly available at: https://github.com/yftian2025/SynUIEDatasets.git.

cs.CV↗