arXiv subjects

Zhihui Li

Publications and source records attributed to Zhihui Li.

At least 37 records · Page 2Linked to original sources

arXiv2026

The Prevalence of Turbulence-Regulated Multiphase Galactic Winds in Star-Forming Galaxies

We build upon our previously developed multi-ion radiative transfer (RT) framework, PEACOCK, to investigate the kinematic and energetic structure of cool-to-warm galactic winds in a sample of 50 nearby star-forming galaxies. Using self-consistent constraints derived from joint modeling of Ly-alpha and multiple ultraviolet metal lines, we analyze how bulk outflows and turbulent motions contribute to the dynamics and energy budget of galactic winds in the circumgalactic medium (CGM). We find that macroscopic turbulent velocities are often comparable to, and sometimes exceed, the coherent bulk outflow velocity. The associated turbulent pressure frequently dominates over both microscopic pressure and ram pressure, indicating that turbulence is a major contributor to the kinetic energy budget of the CGM wind. Wind kinematics, ionic column densities, and metal mass outflow rates all scale systematically with stellar mass and star formation rate, demonstrating a strong coupling between stellar feedback and CGM structure. Including turbulent motions strengthens these CGM-galaxy scaling relations and favors an energy-driven feedback regime. The total kinetic energy flux of the cool-to-warm CGM correlates tightly with the mechanical energy injection rate from star formation, implying that stellar feedback provides sufficient power to sustain both coherent outflows and turbulence. Comparisons with phenomenological line-profile fitting methods further show that simplified treatments can introduce systematic biases in inferred wind properties. Together these results support a turbulence-regulated picture of galactic winds in which a substantial fraction of feedback energy is stored in turbulent motions within a multiphase CGM.

astro-ph.GA↗

arXiv2025

ParticleGS: Learning Neural Gaussian Particle Dynamics from Videos for Prior-free Physical Motion Extrapolation

The ability to extrapolate dynamic 3D scenes beyond the observed timeframe is fundamental to advancing physical world understanding and predictive modeling. Existing dynamic 3D reconstruction methods have achieved high-fidelity rendering of temporal interpolation, but typically lack physical consistency in predicting the future. To overcome this issue, we propose ParticleGS, a physics-based framework that reformulates dynamic 3D scenes as physically grounded systems. ParticleGS comprises three key components: 1) an encoder that decomposes the scene into static properties and initial dynamic physical fields; 2) an evolver based on Neural Ordinary Differential Equations (Neural ODEs) that learns continuous-time dynamics for motion extrapolation; and 3) a decoder that reconstructs 3D Gaussians from evolved particle states for rendering. Through this design, ParticleGS integrates physical reasoning into dynamic 3D representations, enabling accurate and consistent prediction of the future. Experiments show that ParticleGS achieves state-of-the-art performance in extrapolation while maintaining rendering quality comparable to leading dynamic 3D reconstruction methods.

cs.CV↗

arXiv2025

Saliency-R1: Incentivizing Unified Saliency Reasoning Capability in MLLM with Confidence-Guided Reinforcement Learning

Although multimodal large language models (MLLMs) excel in high-level vision-language reasoning, they lack inherent awareness of visual saliency, making it difficult to identify key visual elements. To bridge this gap, we propose Saliency-R1, the first unified MLLM framework that jointly tackles three representative and heterogeneous saliency tasks: Salient Object Detection (SOD), Salient Instance Segmentation (SIS), and Co-salient Object Detection (CoSOD), enhancing the model's capacity for saliency reasoning. We introduce a textual interface with structured tags ( , ) to encode region- and instance-level referring expressions, enabling a single referring segmenter to produce task-appropriate masks. To train the MLLM efficiently, we propose Confidence-Guided Policy Optimization (CGPO), a novel single-sample reinforcement learning algorithm. CGPO improves on GRPO by replacing group-normalized advantages with a per-sample signal based on reward-confidence discrepancy, thereby reducing computational waste, mitigating signal dilution, and lowering training overhead. Our model exceeds or matches the performance of robust open/closed-source MLLMs and specialized state-of-the-art methods across all three tasks, demonstrating the efficacy of our framework in saliency reasoning.

cs.CV↗

arXiv2025

Which Layer Causes Distribution Deviation? Entropy-Guided Adaptive Pruning for Diffusion and Flow Models

Large-scale vision generative models, including diffusion and flow models, have demonstrated remarkable performance in visual generation tasks. However, transferring these pre-trained models to downstream tasks often results in significant parameter redundancy. In this paper, we propose EntPruner, an entropy-guided automatic progressive pruning framework for diffusion and flow models. First, we introduce entropy-guided pruning, a block-level importance assessment strategy specifically designed for generative models. Unlike discriminative models, generative models require preserving the diversity and condition-fidelity of the output distribution. As the importance of each module can vary significantly across downstream tasks, EntPruner prioritizes pruning of less important blocks using data-dependent Conditional Entropy Deviation (CED) as a guiding metric. CED quantifies how much the distribution diverges from the learned conditional data distribution after removing a block. Second, we propose a zero-shot adaptive pruning framework to automatically determine when and how much to prune during training. This dynamic strategy avoids the pitfalls of one-shot pruning, mitigating mode collapse, and preserving model performance. Extensive experiments on DiT and SiT models demonstrate the effectiveness of EntPruner, achieving up to 2.22$\times$ inference speedup while maintaining competitive generation quality on ImageNet and three downstream datasets.

cs.CV↗

arXiv2025

Efficient Training for Human Video Generation with Entropy-Guided Prioritized Progressive Learning

Human video generation has advanced rapidly with the development of diffusion models, but the high computational cost and substantial memory consumption associated with training these models on high-resolution, multi-frame data pose significant challenges. In this paper, we propose Entropy-Guided Prioritized Progressive Learning (Ent-Prog), an efficient training framework tailored for diffusion models on human video generation. First, we introduce Conditional Entropy Inflation (CEI) to assess the importance of different model components on the target conditional generation task, enabling prioritized training of the most critical components. Second, we introduce an adaptive progressive schedule that adaptively increases computational complexity during training by measuring the convergence efficiency. Ent-Prog reduces both training time and GPU memory consumption while maintaining model performance. Extensive experiments across three datasets, demonstrate the effectiveness of Ent-Prog, achieving up to 2.2$\times$ training speedup and 2.4$\times$ GPU memory reduction without compromising generative performance.

cs.CV↗

arXiv2025

Token Painter: Training-Free Text-Guided Image Inpainting via Mask Autoregressive Models

Text-guided image inpainting aims to inpaint masked image regions based on a textual prompt while preserving the background. Although diffusion-based methods have become dominant, their property of modeling the entire image in latent space makes it challenging for the results to align well with prompt details and maintain a consistent background. To address these issues, we explore Mask AutoRegressive (MAR) models for this task. MAR naturally supports image inpainting by generating latent tokens corresponding to mask regions, enabling better local controllability without altering the background. However, directly applying MAR to this task makes the inpainting content either ignore the prompts or be disharmonious with the background context. Through analysis of the attention maps from the inpainting images, we identify the impact of background tokens on text tokens during the MAR generation, and leverage this to design \textbf{Token Painter}, a training-free text-guided image inpainting method based on MAR. Our approach introduces two key components: (1) Dual-Stream Encoder Information Fusion (DEIF), which fuses the semantic and context information from text and background in frequency domain to produce novel guidance tokens, allowing MAR to generate text-faithful inpainting content while keeping harmonious with background context. (2) Adaptive Decoder Attention Score Enhancing (ADAE), which adaptively enhances attention scores on guidance tokens and inpainting tokens to further enhance the alignment of prompt details and the content visual quality. Extensive experiments demonstrate that our training-free method outperforms prior state-of-the-art methods across almost all metrics. Codes: https://github.com/longtaojiang/Token-Painter.

cs.CV↗

arXiv2025

From Ideal to Real: Unified and Data-Efficient Dense Prediction for Real-World Scenarios

Dense prediction tasks hold significant importance of computer vision, aiming to learn pixel-wise annotated labels for input images. Despite advances in this field, existing methods primarily focus on idealized conditions, exhibiting limited real-world generalization and struggling with the acute scarcity of real-world data in practical scenarios. To systematically study this problem, we first introduce DenseWorld, a benchmark spanning a broad set of 25 dense prediction tasks that correspond to urgent real-world applications, featuring unified evaluation across tasks. We then propose DenseDiT, which exploits generative models' visual priors to perform diverse real-world dense prediction tasks through a unified strategy. DenseDiT combines a parameter-reuse mechanism and two lightweight branches that adaptively integrate multi-scale context. This design enables DenseDiT to achieve efficient tuning with less than 0.1% additional parameters, activating the visual priors while effectively adapting to diverse real-world dense prediction tasks. Evaluations on DenseWorld reveal significant performance drops in existing general and specialized baselines, highlighting their limited real-world generalization. In contrast, DenseDiT achieves superior results using less than 0.01% training data of baselines, underscoring its practical value for real-world deployment.

cs.CV↗

arXiv2025

Self-Consistency as a Free Lunch: Reducing Hallucinations in Vision-Language Models via Self-Reflection

Vision-language models often hallucinate details, generating non-existent objects or inaccurate attributes that compromise output reliability. Existing methods typically address these issues via extensive human annotations or external supervision from more powerful models. In this work, we present a novel framework that leverages the model's self-consistency between long responses and short answers to generate preference pairs for training. We observe that short binary questions tend to yield highly reliable responses, which can be used to query the target model to evaluate and rank its generated responses. Specifically, we design a self-reflection pipeline where detailed model responses are compared against concise binary answers, and inconsistency signals are utilized to automatically curate high-quality training data without human annotations or external model-based supervision. By relying solely on self-consistency rather than external supervision, our method offers a scalable and efficient solution that effectively reduces hallucinations using unlabeled data. Extensive experiments on multiple benchmarks, i.e., AMBER, MultiObject-Hal (ROPE), Object HalBench, and MMHal-Bench, demonstrate significant improvements in factual grounding and reliability. Moreover, our approach maintains robust instruction-following ability, as evidenced by enhanced performance on LLaVA-Bench and MMBench.

cs.CV↗

arXiv2025

Mettle: Meta-Token Learning for Memory-Efficient Audio-Visual Adaptation

We present \textbf{Met}a-\textbf{T}oken \textbf{Le}arning (Mettle), a simple and memory-efficient method for adapting large-scale pretrained transformer models to downstream audio-visual tasks. Instead of sequentially modifying the output feature distribution of the transformer backbone, Mettle utilizes a lightweight \textit{Layer-Centric Distillation (LCD)} module to distill in parallel the intact audio or visual features embedded by each transformer layer into compact meta-tokens. This distillation process considers both pretrained knowledge preservation and task-specific adaptation. The obtained meta-tokens can be directly applied to classification tasks, such as audio-visual event localization and audio-visual video parsing. To further support fine-grained segmentation tasks, such as audio-visual segmentation, we introduce a \textit{Meta-Token Injection (MTI)} module, which utilizes the audio and visual meta-tokens distilled from the top transformer layer to guide feature adaptation in earlier layers. Extensive experiments on multiple audiovisual benchmarks demonstrate that our method significantly reduces memory usage and training time while maintaining parameter efficiency and competitive accuracy.

cs.CV↗

arXiv2025

CoNav: Collaborative Cross-Modal Reasoning for Embodied Navigation

Embodied navigation demands comprehensive scene understanding and precise spatial reasoning. While image-text models excel at interpreting pixel-level color and lighting cues, 3D-text models capture volumetric structure and spatial relationships. However, unified fusion approaches that jointly fuse 2D images, 3D point clouds, and textual instructions face challenges in limited availability of triple-modality data and difficulty resolving conflicting beliefs among modalities. In this work, we introduce CoNav, a collaborative cross-modal reasoning framework where a pretrained 3D-text model explicitly guides an image-text navigation agent by providing structured spatial-semantic knowledge to resolve ambiguities during navigation. Specifically, we introduce Cross-Modal Belief Alignment, which operationalizes this cross-modal guidance by simply sharing textual hypotheses from the 3D-text model to the navigation agent. Through lightweight fine-tuning on a small 2D-3D-text corpus, the navigation agent learns to integrate visual cues with spatial-semantic knowledge derived from the 3D-text model, enabling effective reasoning in embodied navigation. CoNav achieves significant improvements on four standard embodied navigation benchmarks (R2R, CVDN, REVERIE, SOON) and two spatial reasoning benchmarks (ScanQA, SQA3D). Moreover, under close navigation Success Rate, CoNav often generates shorter paths compared to other methods (as measured by SPL), showcasing the potential and challenges of fusing data from different modalities in embodied navigation. Project Page: https://oceanhao.github.io/CoNav/

cs.CV↗

arXiv2025

FANeRV: Frequency Separation and Augmentation based Neural Representation for Video

Neural representations for video (NeRV) have gained considerable attention for their strong performance across various video tasks. However, existing NeRV methods often struggle to capture fine spatial details, resulting in vague reconstructions. In this paper, we present a Frequency Separation and Augmentation based Neural Representation for video (FANeRV), which addresses these limitations with its core Wavelet Frequency Upgrade Block. This block explicitly separates input frames into high and low-frequency components using discrete wavelet transform, followed by targeted enhancement using specialized modules. Finally, a specially designed gated network effectively fuses these frequency components for optimal reconstruction. Additionally, convolutional residual enhancement blocks are integrated into the later stages of the network to balance parameter distribution and improve the restoration of high-frequency details. Experimental results demonstrate that FANeRV significantly improves reconstruction performance and excels in multiple tasks, including video compression, inpainting, and interpolation, outperforming existing NeRV methods.

cs.CV↗

arXiv2025

High-Frequency Enhanced Hybrid Neural Representation for Video Compression

Neural Representations for Videos (NeRV) have simplified the video codec process and achieved swift decoding speeds by encoding video content into a neural network, presenting a promising solution for video compression. However, existing work overlooks the crucial issue that videos reconstructed by these methods lack high-frequency details. To address this problem, this paper introduces a High-Frequency Enhanced Hybrid Neural Representation Network. Our method focuses on leveraging high-frequency information to improve the synthesis of fine details by the network. Specifically, we design a wavelet high-frequency encoder that incorporates Wavelet Frequency Decomposer (WFD) blocks to generate high-frequency feature embeddings. Next, we design the High-Frequency Feature Modulation (HFM) block, which leverages the extracted high-frequency embeddings to enhance the fitting process of the decoder. Finally, with the refined Harmonic decoder block and a Dynamic Weighted Frequency Loss, we further reduce the potential loss of high-frequency information. Experiments on the Bunny and UVG datasets demonstrate that our method outperforms other methods, showing notable improvements in detail preservation and compression performance.

cs.CV↗

arXiv2025

Learning A Zero-shot Occupancy Network from Vision Foundation Models via Self-supervised Adaptation

Estimating the 3D world from 2D monocular images is a fundamental yet challenging task due to the labour-intensive nature of 3D annotations. To simplify label acquisition, this work proposes a novel approach that bridges 2D vision foundation models (VFMs) with 3D tasks by decoupling 3D supervision into an ensemble of image-level primitives, e.g., semantic and geometric components. As a key motivator, we leverage the zero-shot capabilities of vision-language models for image semantics. However, due to the notorious ill-posed problem - multiple distinct 3D scenes can produce identical 2D projections, directly inferring metric depth from a monocular image in a zero-shot manner is unsuitable. In contrast, 2D VFMs provide promising sources of relative depth, which theoretically aligns with metric depth when properly scaled and offset. Thus, we adapt the relative depth derived from VFMs into metric depth by optimising the scale and offset using temporal consistency, also known as novel view synthesis, without access to ground-truth metric depth. Consequently, we project the semantics into 3D space using the reconstructed metric depth, thereby providing 3D supervision. Extensive experiments on nuScenes and SemanticKITTI demonstrate the effectiveness of our framework. For instance, the proposed method surpasses the current state-of-the-art by 3.34% mIoU on nuScenes for voxel occupancy prediction.

cs.CV↗

arXiv2025

Machine Learning-based Regional Cooling Demand Prediction with Optimised Dataset Partitioning

In the context of global warming, even relatively cooler countries like the UK are experiencing a rise in cooling demand, particularly in southern regions such as London. This growing demand, especially during the summer months, presents significant challenges for energy management systems. Accurately predicting cooling demand in urban domestic buildings is essential for maintaining energy efficiency. This study introduces a generalised framework for developing high-resolution Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) networks using physical model-based summer cooling demand data. To maximise the predictive capability and generalisation ability of the models under limited data scenarios, four distinct data partitioning strategies were implemented, including the extrapolation, month-based interpolation, global interpolation, and day-based interpolation. Bayesian Optimisation (BO) was then applied to fine-tune the hyper-parameters, substantially improving the framework predictive accuracy. Results show that the day-based interpolation GRU model demonstrated the best performance due to its ability to retain both the data randomness and the time sequence continuity characteristics. This optimal model achieves a Root Mean Squared Error (RMSE) of 2.22%, a Mean Absolute Error (MAE) of 0.87%, and a coefficient of determination (R square) of 0.9386 on the test set. The generalisation ability of this framework was further evaluated by forecasting.

physics.soc-ph↗

arXiv2025

Gas Accretion Traced by Blue-Dominated Ly$α$ Emission and Redshifted UV Absorption Lines in Protocluster Galaxies at z = 2.3 from the KBSS-KCWI Survey

Ly$α$ emission with a dominant blueshifted peak can probe gas flowing through the circumgalactic medium as it accretes onto galaxies and fuels new star formation, although it has seldom actually been observed. Here we present new Keck Cosmic Web Imager observations of the extended Ly$α$ halos surrounding Q1700-BX710 and Q1700-BX711, a pair of UV continuum-selected Keck Baryonic Structure Survey (KBSS) galaxies at z = 2.3 in the HS1700+643 protocluster. We find that BX710 and BX711's Ly$α$ halos are aligned with a large-scale galaxy filament consisting of thirteen spectroscopically identified protocluster galaxies. By measuring the peak separation and blue-to-red peak flux ratio of the Ly$α$ emission profiles throughout these galaxies' Ly$α$ halos, we have obtained measurements of their spatially varying velocity structure. The prevalence of blue-dominated Ly$α$ emission profiles throughout BX711's Ly$α$ halo suggests actively accreting gas. We fit a clumpy, multiphase Monte Carlo Radiative Transfer model which assumes a radially varying clump velocity to the spatially resolved Ly$α$ emission throughout BX710 and BX711's Ly$α$ halos and simultaneously fit these galaxies' average down-the-barrel UV absorption profile with a radially varying velocity model. The results of these models are consistent with a combination of HI and higher-metallicity gas accretion for both galaxies, especially BX711, which exhibits inflow-driven kinematics throughout most of its Ly$α$ halo. We consider various accretion scenarios to explain these findings, including accretion of metal-enriched gas from the cosmic web, galaxy interactions, and recycled gas from the circumgalactic medium, all of which are compatible with our current observations.

astro-ph.GA↗

arXiv2024

Unsupervised Visible-Infrared Person ReID by Collaborative Learning with Neighbor-Guided Label Refinement

Unsupervised learning visible-infrared person re-identification (USL-VI-ReID) aims at learning modality-invariant features from unlabeled cross-modality dataset, which is crucial for practical applications in video surveillance systems. The key to essentially address the USL-VI-ReID task is to solve the cross-modality data association problem for further heterogeneous joint learning. To address this issue, we propose a Dual Optimal Transport Label Assignment (DOTLA) framework to simultaneously assign the generated labels from one modality to its counterpart modality. The proposed DOTLA mechanism formulates a mutual reinforcement and efficient solution to cross-modality data association, which could effectively reduce the side-effects of some insufficient and noisy label associations. Besides, we further propose a cross-modality neighbor consistency guided label refinement and regularization module, to eliminate the negative effects brought by the inaccurate supervised signals, under the assumption that the prediction or label distribution of each example should be similar to its nearest neighbors. Extensive experimental results on the public SYSU-MM01 and RegDB datasets demonstrate the effectiveness of the proposed method, surpassing existing state-of-the-art approach by a large margin of 7.76% mAP on average, which even surpasses some supervised VI-ReID methods.

cs.CV↗

arXiv2024

Synergistic Radiative Transfer Modeling of MgII and Lyα Emission in Multiphase, Clumpy Galactic Environments: Application to Low-Redshift Lyman Continuum Leakers

We conducted systematic radiative transfer (RT) modeling of the Mg II doublet line profiles for 33 low-redshift Lyman continuum (LyC) leakers, and Ly$α$ modeling for a subset of six objects, using a multiphase, clumpy circumgalactic medium (CGM) model. Our RT models successfully reproduced the Mg II line profiles for all 33 galaxies, revealing a necessary condition for strong LyC leakage: high maximum clump outflow velocity ($v_{\rm MgII,\,max} \gtrsim 390\,\rm km\,s^{-1}$) and low total Mg II column density ($N_{\rm MgII,\,tot} \lesssim 10^{14.3}\,\rm cm^{-2}$). We found that the clump outflow velocity and total Mg II column density have the most significant impact on Mg II spectra and emphasized the need for full RT modeling to accurately extract the CGM gas properties. In addition, using archival HST COS/G160M data, we modeled Ly$α$ profiles for six objects and found that their spectral properties do not fully align with the conventional LyC leakage criteria, yet no clear correlation was identified between the modeled parameters and observed LyC escape fractions. We inferred LyC escape fractions based on HI properties from Ly$α$ RT modeling and found that LyC leakage is primarily governed by the number of optically thick HI clumps per sightline ($f_{\rm cl}$). Intriguingly, two galaxies with relatively low observed LyC leakage exhibited the highest RT-inferred LyC escape fractions due to their lowest $f_{\rm cl}$ values, driven by the strong blue peaks of their Ly$α$ emission. Future high-resolution, spatially resolved observations are crucial for resolving this puzzle. Overall, our results support a "picket fence" geometry over a "density-bounded" scenario for the CGM, where a combination of high Mg II outflow velocities and low Mg II column densities may be correlated with the presence of more low-density HI channels that facilitate LyC escape.

astro-ph.GA↗

arXiv2024

Ly$α$ Emission Line Profiles of Extreme [OIII] Emitting Galaxies at $z\gtrsim2$: Implications for Ly$α$ Visibility in the Reionization Era

JWST observations have recently begun delivering the first samples of Ly$α$ velocity profile measurements at $z>6$, opening a new window on the reionization process. Interpretation of $z\gtrsim6$ line profiles is currently stunted by limitations in our knowledge of the intrinsic Ly$α$ profile (before encountering the IGM) of the galaxies that are common at $z\gtrsim6$. To overcome this shortcoming, we have obtained resolved ($R\sim3900$) Ly$α$ spectroscopy of $42$ galaxies at $z=2.1-3.4$ with similar properties as are seen at $z>6$. We quantify a variety of Ly$α$ profile statistics as a function of [OIII]+H$β$ EW. Our spectra reveal a new population of $z\simeq 2-3$ galaxies with large [OIII]+H$β$ EWs ($>1200\ Å$) and a large fraction of Ly$α$ flux emerging near the systemic redshift (peak velocity $\simeq0$ km s$^{-1}$). These spectra indicate that low density neutral hydrogen channels are able to form in a subset of low mass galaxies ($\lesssim1\times10^8\ M_{\odot}$) that experience a burst of star formation (sSFR $>100$ Gyr$^{-1}$). Other extreme [OIII] emitters show weaker Ly$α$ that is shifted to higher velocities ($\simeq240$ km s$^{-1}$) with little emission near line center. We investigate the impact the IGM is likely to have on these intrinsic line profiles in the reionization era, finding that the centrally peaked Ly$α$ emitters should be strongly attenuated at $z\gtrsim5$. We show that these line profiles are particularly sensitive to the impact of resonant scattering from infalling IGM and can be strongly attenuated even when the IGM is highly ionized at $z\simeq 5$. We compare these expectations against a new database of $z\gtrsim6.5$ galaxies with robust velocity profiles measured with JWST/NIRSpec.

astro-ph.GA↗