arXiv subjects

Jiang Liu

Publications and source records attributed to Jiang Liu.

At least 73 records · Page 4Linked to original sources

arXiv2025

TECM*: A Data-Driven Assessment to Reinforcement Learning Methods and Application to Heparin Treatment Strategy for Surgical Sepsis

Objective: Sepsis is a life-threatening condition caused by severe infection leading to acute organ dysfunction. This study proposes a data-driven metric and a continuous reward function to optimize personalized heparin therapy in surgical sepsis patients. Methods: Data from the MIMIC-IV v1.0 and eICU v2.0 databases were used for model development and evaluation. The training cohort consisted of abdominal surgery patients receiving unfractionated heparin (UFH) after postoperative sepsis onset. We introduce a new RL-based framework: converting the discrete SOFA score to a continuous cxSOFA for more nuanced state and reward functions; Second, defining "good" or "bad" strategies based on cxSOFA by a stepwise manner; Third, proposing a Treatment Effect Comparison Matrix (TECM), analogous to a confusion matrix for classification tasks, to evaluate the treatment strategies. We applied different RL algorithms, Q-Learning, DQN, DDQN, BCQ and CQL to optimize the treatment and comprehensively evaluated the framework. Results: Among the AI-derived strategies, the cxSOFA-CQL model achieved the best performance, reducing mortality from 1.83% to 0.74% with the average hospital stay from 11.11 to 9.42 days. TECM demonstrated consistent outcomes across models, highlighting robustness. Conclusion: The proposed RL framework enables interpretable and robust optimization of heparin therapy in surgical sepsis. Continuous cxSOFA scoring and TECM-based evaluation provide nuanced treatment assessment, showing promise for improving clinical outcomes and decision-support reliability.

cs.LG↗

arXiv2025

Unleashing Hour-Scale Video Training for Long Video-Language Understanding

Recent long-form video-language understanding benchmarks have driven progress in video large multimodal models (Video-LMMs). However, the scarcity of well-annotated long videos has left the training of hour-long Video-LMMs underexplored. To close this gap, we present VideoMarathon, a large-scale hour-long video instruction-following dataset. This dataset includes around 9,700 hours of long videos sourced from diverse domains, ranging from 3 to 60 minutes per video. Specifically, it contains 3.3M high-quality QA pairs, spanning six fundamental topics: temporality, spatiality, object, action, scene, and event. Compared to existing video instruction datasets, VideoMarathon significantly extends training video durations up to 1 hour, and supports 22 diverse tasks requiring both short- and long-term video comprehension. Building on VideoMarathon, we propose Hour-LLaVA, a powerful and efficient Video-LMM for hour-scale video-language modeling. It enables hour-long video training and inference at 1-FPS sampling by leveraging a memory augmentation module, which adaptively integrates question-relevant and spatiotemporally informative semantics from the cached full video context. In our experiments, Hour-LLaVA achieves the best performance on multiple representative long video-language benchmarks, demonstrating the high quality of the VideoMarathon dataset and the superiority of the Hour-LLaVA model.

cs.CV↗

arXiv2025

DiffProtect: Generate Adversarial Examples with Diffusion Models for Facial Privacy Protection

The increasingly pervasive facial recognition (FR) systems raise serious concerns about personal privacy, especially for billions of users who have publicly shared their photos on social media. Several attempts have been made to protect individuals from being identified by unauthorized FR systems utilizing adversarial attacks to generate encrypted face images. However, existing methods suffer from poor visual quality or low attack success rates, which limit their utility. Recently, diffusion models have achieved tremendous success in image generation. In this work, we ask: can diffusion models be used to generate adversarial examples to improve both visual quality and attack performance? We propose DiffProtect, which utilizes a diffusion autoencoder to generate semantically meaningful perturbations on FR systems. Extensive experiments demonstrate that DiffProtect produces more natural-looking encrypted images than state-of-the-art methods while achieving significantly higher attack success rates, e.g., 24.5% and 25.1% absolute improvements on the CelebA-HQ and FFHQ datasets.

cs.CV↗

arXiv2025

DeLightMono: Enhancing Self-Supervised Monocular Depth Estimation in Endoscopy by Decoupling Uneven Illumination

Self-supervised monocular depth estimation serves as a key task in the development of endoscopic navigation systems. However, performance degradation persists due to uneven illumination inherent in endoscopic images, particularly in low-intensity regions. Existing low-light enhancement techniques fail to effectively guide the depth network. Furthermore, solutions from other fields, like autonomous driving, require well-lit images, making them unsuitable and increasing data collection burdens. To this end, we present DeLight-Mono - a novel self-supervised monocular depth estimation framework with illumination decoupling. Specifically, endoscopic images are represented by a designed illumination-reflectance-depth model, and are decomposed with auxiliary networks. Moreover, a self-supervised joint-optimizing framework with novel losses leveraging the decoupled components is proposed to mitigate the effects of uneven illumination on depth estimation. The effectiveness of the proposed methods was rigorously verified through extensive comparisons and an ablation study performed on two public datasets.

cs.CV↗

arXiv2025

Instella: Fully Open Language Models with Stellar Performance

Large language models (LLMs) have demonstrated remarkable performance across a wide range of tasks, yet the majority of high-performing models remain closed-source or partially open, limiting transparency and reproducibility. In this work, we introduce Instella, a family of fully open three billion parameter language models trained entirely on openly available data and codebase. Powered by AMD Instinct MI300X GPUs, Instella is developed through large-scale pre-training, general-purpose instruction tuning, and alignment with human preferences. Despite using substantially fewer pre-training tokens than many contemporaries, Instella achieves state-of-the-art results among fully open models and is competitive with leading open-weight models of comparable size. We further release two specialized variants: Instella-Long, capable of handling context lengths up to 128K tokens, and Instella-Math, a reasoning-focused model enhanced through supervised fine-tuning and reinforcement learning on mathematical tasks. Together, these contributions establish Instella as a transparent, performant, and versatile alternative for the community, advancing the goal of open and reproducible language modeling research.

cs.CL↗

arXiv2025

Learning from Online Videos at Inference Time for Computer-Use Agents

Computer-use agents can operate computers and automate laborious tasks, but despite recent rapid progress, they still lag behind human users, especially when tasks require domain-specific procedural knowledge about particular applications, platforms, and multi-step workflows. Humans can bridge this gap by watching video tutorials: we search, skim, and selectively imitate short segments that match our current subgoal. In this paper, we study how to enable computer-use agents to learn from online videos at inference time effectively. We propose a framework that retrieves and filters tutorial videos, converts them into structured demonstration trajectories, and dynamically selects trajectories as in-context guidance during execution. Particularly, using a VLM, we infer UI actions, segment videos into short subsequences of actions, and assign each subsequence a textual objective. At inference time, a two-stage selection mechanism dynamically chooses a single trajectory to add in context at each step, focusing the agent on the most helpful local guidance for its next decision. Experiments on two widely used benchmarks show that our framework consistently outperforms strong base agents and variants that use only textual tutorials or transcripts. Analyses highlight the importance of trajectory segmentation and selection, action filtering, and visual information, suggesting that abundant online videos can be systematically distilled into actionable guidance that improves computer-use agents at inference time. Our code is available at https://github.com/UCSB-NLP-Chang/video_demo.

cs.CV↗

arXiv2025

Deep Q-Network for Optimizing NOMA-Aided Resource Allocation in Smart Factories with URLLC Constraints

This paper presents a Deep Q-Network (DQN)- based algorithm for NOMA-aided resource allocation in smart factories, addressing the stringent requirements of Ultra-Reliable Low-Latency Communication (URLLC). The proposed algorithm dynamically allocates sub-channels and optimizes power levels to maximize throughput while meeting strict latency constraints. By incorporating a tunable parameter λ, the algorithm balances the trade-off between throughput and latency, making it suitable for various devices, including robots, sensors, and controllers, each with distinct communication needs. Simulation results show that robots achieve higher throughput, while sensors and controllers meet the low-latency requirements of URLLC, ensuring reliable communication for real-time industrial applications.

eess.SY↗

arXiv2025

Cross-Modal Scene Semantic Alignment for Image Complexity Assessment

Image complexity assessment (ICA) is a challenging task in perceptual evaluation due to the subjective nature of human perception and the inherent semantic diversity in real-world images. Existing ICA methods predominantly rely on hand-crafted or shallow convolutional neural network-based features of a single visual modality, which are insufficient to fully capture the perceived representations closely related to image complexity. Recently, cross-modal scene semantic information has been shown to play a crucial role in various computer vision tasks, particularly those involving perceptual understanding. However, the exploration of cross-modal scene semantic information in the context of ICA remains unaddressed. Therefore, in this paper, we propose a novel ICA method called Cross-Modal Scene Semantic Alignment (CM-SSA), which leverages scene semantic alignment from a cross-modal perspective to enhance ICA performance, enabling complexity predictions to be more consistent with subjective human perception. Specifically, the proposed CM-SSA consists of a complexity regression branch and a scene semantic alignment branch. The complexity regression branch estimates image complexity levels under the guidance of the scene semantic alignment branch, while the scene semantic alignment branch is used to align images with corresponding text prompts that convey rich scene semantic information by pair-wise learning. Extensive experiments on several ICA datasets demonstrate that the proposed CM-SSA significantly outperforms state-of-the-art approaches. Codes are available at https://github.com/XQ2K/First-Cross-Model-ICA.

cs.CV↗

arXiv2025

KeyVID: Keyframe-Aware Video Diffusion for Audio-Synchronized Visual Animation

Generating video from various conditions, such as text, image, and audio, enables both spatial and temporal control, leading to high-quality generation results. Videos with dramatic motions often require a higher frame rate to ensure smooth motion. Currently, most audio-to-visual animation models use uniformly sampled frames from video clips. However, these uniformly sampled frames fail to capture significant key moments in dramatic motions at low frame rates and require significantly more memory when increasing the number of frames directly. In this paper, we propose KeyVID, a keyframe-aware audio-to-visual animation framework that significantly improves the generation quality for key moments in audio signals while maintaining computation efficiency. Given an image and an audio input, we first localize keyframe time steps from the audio. Then, we use a keyframe generator to generate the corresponding visual keyframes. Finally, we generate all intermediate frames using the motion interpolator. Through extensive experiments, we demonstrate that KeyVID significantly improves audio-video synchronization and video quality across multiple datasets, particularly for highly dynamic motions. The code is released in https://github.com/XingruiWang/KeyVID.

cs.CV↗

arXiv2025

Mesh-Gait: A Unified Framework for Gait Recognition Through Multi-Modal Representation Learning from 2D Silhouettes

Gait recognition, a fundamental biometric technology, leverages unique walking patterns for individual identification, typically using 2D representations such as silhouettes or skeletons. However, these methods often struggle with viewpoint variations, occlusions, and noise. Multi-modal approaches that incorporate 3D body shape information offer improved robustness but are computationally expensive, limiting their feasibility for real-time applications. To address these challenges, we introduce Mesh-Gait, a novel end-to-end multi-modal gait recognition framework that directly reconstructs 3D representations from 2D silhouettes, effectively combining the strengths of both modalities. Compared to existing methods, directly learning 3D features from 3D joints or meshes is complex and difficult to fuse with silhouette-based gait features. To overcome this, Mesh-Gait reconstructs 3D heatmaps as an intermediate representation, enabling the model to effectively capture 3D geometric information while maintaining simplicity and computational efficiency. During training, the intermediate 3D heatmaps are gradually reconstructed and become increasingly accurate under supervised learning, where the loss is calculated between the reconstructed 3D joints, virtual markers, and 3D meshes and their corresponding ground truth, ensuring precise spatial alignment and consistent 3D structure. Mesh-Gait extracts discriminative features from both silhouettes and reconstructed 3D heatmaps in a computationally efficient manner. This design enables the model to capture spatial and structural gait characteristics while avoiding the heavy overhead of direct 3D reconstruction from RGB videos, allowing the network to focus on motion dynamics rather than irrelevant visual details. Extensive experiments demonstrate that Mesh-Gait achieves state-of-the-art accuracy. The code will be released upon acceptance of the paper.

cs.CV↗

arXiv2025

Latent Visual Reasoning

Multimodal Large Language Models (MLLMs) have achieved notable gains in various tasks by incorporating Chain-of-Thought (CoT) reasoning in language spaces. Recent work extends this direction by leveraging external tools for visual editing, thereby enhancing the visual signal along the reasoning trajectories. Nevertheless, these approaches remain fundamentally constrained: reasoning is still confined to the language space, with visual information treated as static preconditions. We introduce Latent Visual Reasoning (LVR), a new paradigm that enables autoregressive reasoning directly in the visual embedding space. A visual encoder first projects images into visual tokens within a joint semantic space shared with the language model. The language model is then trained to generate latent states that reconstruct key visual tokens critical for answering the query, constituting the process of latent visual reasoning. By interleaving LVR with standard text generation, our model achieves substantial gains on perception-intensive visual question answering tasks. In addition, we adapt the GRPO algorithm to conduct reinforcement learning on latent reasoning, further balancing LVR and textual generation. We show that LVR substantially improves fine-grained visual understanding and perception, achieving 71.67% on MMVP compared to 66.67% with Qwen2.5-VL. Code base and model weights will be released later.

cs.CV↗

arXiv2025

ImageDoctor: Diagnosing Text-to-Image Generation via Grounded Image Reasoning

The rapid advancement of text-to-image (T2I) models has increased the need for reliable human preference modeling, a demand further amplified by recent progress in reinforcement learning for preference alignment. However, existing approaches typically quantify the quality of a generated image using a single scalar, limiting their ability to provide comprehensive and interpretable feedback on image quality. To address this, we introduce ImageDoctor, a unified multi-aspect T2I model evaluation framework that assesses image quality across four complementary dimensions: plausibility, semantic alignment, aesthetics, and overall quality. ImageDoctor also provides pixel-level flaw indicators in the form of heatmaps, which highlight misaligned or implausible regions, and can be used as a dense reward for T2I model preference alignment. Inspired by the diagnostic process, we improve the detail sensitivity and reasoning capability of ImageDoctor by introducing a "look-think-predict" paradigm, where the model first localizes potential flaws, then generates reasoning, and finally concludes the evaluation with quantitative scores. Built on top of a vision-language model and trained through a combination of supervised fine-tuning and reinforcement learning, ImageDoctor demonstrates strong alignment with human preference across multiple datasets, establishing its effectiveness as an evaluation metric. Furthermore, when used as a reward model for preference tuning, ImageDoctor significantly improves generation quality -- achieving an improvement of 10% over scalar-based reward models.

cs.CV↗

arXiv2025

APRIL: Active Partial Rollouts in Reinforcement Learning to Tame Long-tail Generation

Reinforcement learning (RL) has become a cornerstone in advancing large-scale pre-trained language models (LLMs). Successive generations, including GPT-o series, DeepSeek-R1, Kimi-K1.5, Grok 4, and GLM-4.5, have relied on large-scale RL training to enhance reasoning and coding capabilities. To meet the community's growing RL needs, numerous RL frameworks have been proposed. However, RL training remains computationally expensive, with rollout generation accounting for more than 90% of total runtime. In addition, its efficiency is often constrained by the long-tail distribution of rollout response lengths, where a few lengthy responses stall entire batches, leaving GPUs idle and underutilized. As model and rollout sizes continue to grow, this bottleneck increasingly limits scalability. To address this challenge, we propose Active Partial Rollouts in Reinforcement Learning (APRIL), which mitigates long-tail inefficiency. In the rollout phase, APRIL over-provisions rollout requests, terminates once the target number of responses is reached, and recycles incomplete responses for continuation in future steps. This strategy ensures that no rollouts are discarded while substantially reducing GPU idle time. Experiments show that APRIL improves rollout throughput by 22.5% on average (at most 44%) across commonly used RL algorithms (GRPO, DAPO, GSPO), accelerates convergence, and achieves 2.1% on average(at most 8%) higher final accuracy across tasks. Moreover, APRIL is both framework and hardware agnostic, already integrated into the slime RL framework, and deployable on NVIDIA and AMD GPUs alike. Taken together, this work unifies system-level and algorithmic considerations in proposing APRIL, with the aim of advancing RL training efficiency and inspiring further optimizations in RL systems. Our codebase is available at https://github.com/RLsys-Foundation/APRIL

cs.LG↗

arXiv2025

Adaptive Confidence-Wise Loss for Improved Lens Structure Segmentation in AS-OCT

Precise lens structure segmentation is essential for the design of intraocular lenses (IOLs) in cataract surgery. Existing deep segmentation networks typically weight all pixels equally under cross-entropy (CE) loss, overlooking the fact that sub-regions of lens structures are inhomogeneous (e.g., some regions perform better than others) and that boundary regions often suffer from poor segmentation calibration at the pixel level. Clinically, experts annotate different sub-regions of lens structures with varying confidence levels, considering factors such as sub-region proportions, ambiguous boundaries, and lens structure shapes. Motivated by this observation, we propose an Adaptive Confidence-Wise (ACW) loss to group each lens structure sub-region into different confidence sub-regions via a confidence threshold from the unique region aspect, aiming to exploit the potential of expert annotation confidence prior. Specifically, ACW clusters each target region into low-confidence and high-confidence groups and then applies a region-weighted loss to reweigh each confidence group. Moreover, we design an adaptive confidence threshold optimization algorithm to adjust the confidence threshold of ACW dynamically. Additionally, to better quantify the miscalibration errors in boundary region segmentation, we propose a new metric, termed Boundary Expected Calibration Error (BECE). Extensive experiments on a clinical lens structure AS-OCT dataset and other multi-structure datasets demonstrate that our ACW significantly outperforms competitive segmentation loss methods across different deep segmentation networks (e.g., MedSAM). Notably, our method surpasses CE with 6.13% IoU gain, 4.33% DSC increase, and 4.79% BECE reduction in lens structure segmentation under U-Net. The code of this paper is available at https://github.com/XiaoLing12138/Adaptive-Confidence-Wise-Loss.

cs.CV↗

arXiv2025

PAE MobiLLM: Privacy-Aware and Efficient LLM Fine-Tuning on the Mobile Device via Additive Side-Tuning

There is a huge gap between numerous intriguing applications fostered by on-device large language model (LLM) fine-tuning (FT) from fresh mobile data and the limited resources of a mobile device. While existing server-assisted methods (e.g., split learning or side-tuning) may enable LLM FT on the local mobile device, they suffer from heavy communication burdens of activation transmissions, and may disclose data and labels to the server. To address those issues, we develop PAE MobiLLM, a a privacy-aware and efficient LLM FT method which can be deployed on the mobile device via server-assisted additive side-tuning. To further accelerate FT convergence and improve computing efficiency, PAE MobiLLM integrates activation caching on the server side, which allows the server to reuse historical activations and saves the mobile device from repeatedly computing forward passes for the recurring data samples. Besides, to reduce communication cost, PAE MobiLLM develops an activation shortcut that transmits only the token involved in the loss calculation instead of full activation matrices to guide the side network tuning. Last but not least, PAE MobiLLM introduces the additive adapter side-network design which makes the server train the adapter modules based on device-defined prediction differences rather than raw ground-truth labels. In this way, the server can only assist device-defined side-network computing, and learn nothing about data and labels. Extensive experimental results demonstrate PAE MobiLLM's superiority.

cs.LG↗

arXiv2025

Fed MobiLLM: Efficient Federated LLM Fine-Tuning over Heterogeneous Mobile Devices via Server Assisted Side-Tuning

Collaboratively fine-tuning (FT) large language models (LLMs) over heterogeneous mobile devices fosters immense potential applications of personalized intelligence. However, such a vision faces critical system challenges. Conventional federated LLM FT approaches place prohibitive computational and memory burdens on mobile hardware, and their synchronous model aggregation protocols stall for slower devices. In this paper, we propose Fed MobiLLM, a novel design to facilitate efficient federated LLM FT across mobile devices with diverse computing/communication speeds and local model architectures. In particular, Fed MobiLLM implements a pioneering server-assisted federated side-tuning paradigm. Briefly, mobile devices perform lightweight forward propagation computations on local data using their frozen pre-scaled backbone LLMs, and then upload selected intermediate activations. The server trains a shared side-network independently, eliminating client-side backpropagation and enabling asynchronous updates. To bridge model heterogeneity across different devices, we introduce an adaptive layer-wise feature alignment method, which ensures consistent representations for collaboratively tuning a shared side network. Extensive experimental results demonstrate that Fed MobiLLM can maintain robust fine-tuning performance while achieving extremely low on-device memory, with at least 95.2% reduction in computation overhead, 93.2% reduction in communication costs and 5.1x faster convergence compared to existing methods, validating its efficacy for practical LLM adaptation over heterogeneous mobile devices.

cs.LG↗

arXiv2025

Expert-Like Reparameterization of Heterogeneous Pyramid Receptive Fields in Efficient CNNs for Fair Medical Image Classification

Efficient convolutional neural network (CNN) architecture design has attracted growing research interests. However, they typically apply single receptive field (RF), small asymmetric RFs, or pyramid RFs to learn different feature representations, still encountering two significant challenges in medical image classification tasks: 1) They have limitations in capturing diverse lesion characteristics efficiently, e.g., tiny, coordination, small and salient, which have unique roles on the classification results, especially imbalanced medical image classification. 2) The predictions generated by those CNNs are often unfair/biased, bringing a high risk when employing them to real-world medical diagnosis conditions. To tackle these issues, we develop a new concept, Expert-Like Reparameterization of Heterogeneous Pyramid Receptive Fields (ERoHPRF), to simultaneously boost medical image classification performance and fairness. This concept aims to mimic the multi-expert consultation mode by applying the well-designed heterogeneous pyramid RF bag to capture lesion characteristics with varying significances effectively via convolution operations with multiple heterogeneous kernel sizes. Additionally, ERoHPRF introduces an expert-like structural reparameterization technique to merge its parameters with the two-stage strategy, ensuring competitive computation cost and inference speed through comparisons to a single RF. To manifest the effectiveness and generalization ability of ERoHPRF, we incorporate it into mainstream efficient CNN architectures. The extensive experiments show that our proposed ERoHPRF maintains a better trade-off than state-of-the-art methods in terms of medical image classification, fairness, and computation overhead. The code of this paper is available at https://github.com/XiaoLing12138/Expert-Like-Reparameterization-of-Heterogeneous-Pyramid-Receptive-Fields.

cs.CV↗

arXiv2025

PEMUTA: Pedagogically-Enriched Multi-Granular Undergraduate Thesis Assessment

The undergraduate thesis (UGTE) plays an indispensable role in assessing a student's cumulative academic development throughout their college years. Although large language models (LLMs) have advanced education intelligence, they typically focus on holistic assessment with only one single evaluation score, but ignore the intricate nuances across multifaceted criteria, limiting their ability to reflect structural criteria, pedagogical objectives, and diverse academic competencies. Meanwhile, pedagogical theories have long informed manual UGTE evaluation through multi-dimensional assessment of cognitive development, disciplinary thinking, and academic performance, yet remain underutilized in automated settings. Motivated by the research gap, we pioneer PEMUTA, a pedagogically-enriched framework that effectively activates domain-specific knowledge from LLMs for multi-granular UGTE assessment. Guided by Vygotsky's theory and Bloom's Taxonomy, PEMUTA incorporates a hierarchical prompting scheme that evaluates UGTEs across six fine-grained dimensions: Structure, Logic, Originality, Writing, Proficiency, and Rigor (SLOWPR), followed by holistic synthesis. Two in-context learning techniques, \ie, few-shot prompting and role-play prompting, are also incorporated to further enhance alignment with expert judgments without fine-tuning. We curate a dataset of authentic UGTEs with expert-provided SLOWPR-aligned annotations to support multi-granular UGTE assessment. Extensive experiments demonstrate that PEMUTA achieves strong alignment with expert evaluations, and exhibits strong potential for fine-grained, pedagogically-informed UGTE evaluations.

cs.CY↗