Pengcheng Luo

Zhengyao Fang, Zexi Jia, Yijia Zhong et al.

Recent advances in text-to-image (T2I) generation have greatly improved visual quality, yet producing images that appear visually authentic to real-world photography remains challenging. This is partly due to biases in existing evaluation paradigms: human ratings and preference-trained metrics often favor visually vivid images with exaggerated saturation and contrast, which make generations often too vivid to be real even when prompted for realistic-style images. To address this issue, we present Color Fidelity Dataset (CFD) and Color Fidelity Metric (CFM) for objective evaluation of color fidelity in realistic-style generations. CFD contains over 1.3M real and synthetic images with ordered levels of color realism, while CFM employs a multimodal encoder to learn perceptual color fidelity. In addition, we propose a training-free Color Fidelity Refinement (CFR) that adaptively modulates spatial-temporal guidance scale in generation, thereby enhancing color authenticity. Together, CFD supports CFM for assessment, whose learned attention further guides CFR to refine T2I fidelity, forming a progressive framework for assessing and improving color fidelity in realistic-style T2I generation. The dataset and code are available at https://github.com/ZhengyaoFang/CFM.

Hongjun Xie, Jiahang Zhu, Zhiming Shao et al.

Starlink, as a representative low Earth orbit (LEO) satellite broadband system, makes high-bitrate video streaming possible in regions where terrestrial broadband is unavailable. However, its access links exhibit rapid throughput fluctuations caused by satellite mobility and handovers. Existing learned adaptive bitrate (ABR) algorithms can achieve high average quality of experience (QoE), yet high-bitrate Starlink streaming exposes severe session-level rebuffering that is not captured by average QoE alone. To address it, this paper proposes SafeSABR, a risk-calibrated learned ABR framework for Starlink networks. SafeSABR formulates Starlink ABR as a QoE--severe-risk tradeoff and follows a three-stage design: behavior-cloning pretraining learns a high-QoE ABR prior, risk-calibrated reinforcement learning (RL) fine-tuning reduces severe-tail action tendencies, and a runtime safety auditor uses safe-capacity lower bounds to check policy-requested bitrates before execution. Experiments on real Starlink traces compare SafeSABR with online, prediction-assisted, and learned ABR baselines. Compared with advanced methods, SafeSABR reduces severe-stall sessions from 22.8% to 7.2% and worst-5% session rebuffering from 54.30 s to 22.68 s, with a 1.8% QoE cost. Component analyses further show that risk-calibrated fine-tuning and safe-capacity auditing reduce unsafe bitrate decisions and downstream severe-session rebuffering. These results show that combining risk-calibrated policy learning with decision-aware safe throughput forecasting can move learned ABR toward a safer QoE--severe-risk operating point under volatile Starlink networks.

Hongjun Xie, Chao Zhang, Pengcheng Luo et al.

As a representative low Earth orbit (LEO) broadband system, Starlink exhibits highly variable access throughput, making short-term forecasting essential for network resource management. Existing forecasting methods mainly optimize symmetric point-prediction metrics such as MAE and RMSE, but they do not explicitly control the asymmetric risk of overestimating future throughput, which can cause over-admission, bandwidth overbooking, and service violations. This paper formulates Starlink throughput prediction as a risk-budgeted safe forecasting problem, where the predictor must satisfy a prescribed overestimation budget while maintaining competitive accuracy. We propose Budget-Guided Coarse-to-Fine Quantile Selection (BG-CFQS), a data-driven framework that trains a family of lower-quantile predictors, locates the quantile boundary satisfying the risk budget, and refines the boundary region to select the most accurate feasible predictor. Experiments on three real-world Starlink throughput datasets show that BG-CFQS satisfies the risk budget on all datasets and achieves the lowest average MAE, mean positive error, and tail positive error among budget-feasible methods. In high-risk and severe-risk low-throughput regimes, BG-CFQS reduces harmful positive errors by 11.0% and 12.6%, respectively. An admission-control evaluation further shows that the proposed safe forecasts reduce dropped sessions, demonstrating that risk-aware forecasting can translate prediction safety into application-level benefits.

Zexi Jia, Pengcheng Luo, Zhengyao Fang et al.

Classifier-Free Guidance (CFG) serves as the de facto control mechanism for conditional diffusion, yet high guidance scales notoriously induce oversaturation, texture artifacts, and structural collapse. We attribute this failure to a geometric mismatch: standard CFG performs Euclidean extrapolation in ambient space, inadvertently driving sampling trajectories off the high-density data manifold. To resolve this, we present Manifold-Optimal Guidance (MOG), a framework that reformulates guidance as a local optimal control problem. MOG yields a closed-form, geometry-aware Riemannian update that corrects off-manifold drift without requiring retraining. Leveraging this perspective, we further introduce Auto-MOG, a dynamic energy-balancing schedule that adaptively calibrates guidance strength, effectively eliminating the need for manual hyperparameter tuning. Extensive validation demonstrates that MOG yields superior fidelity and alignment compared to baselines, with virtually no added computational overhead.

Lingzi Hong, Pengcheng Luo, Eduardo Blanco et al.

Automatic counterspeech generation methods have been developed to assist efforts in combating hate speech. Existing research focuses on generating counterspeech with linguistic attributes such as being polite, informative, and intent-driven. However, the real impact of counterspeech in online environments is seldom considered. This study aims to develop methods for generating counterspeech constrained by conversation outcomes and evaluate their effectiveness. We experiment with large language models (LLMs) to incorporate into the text generation process two desired conversation outcomes: low conversation incivility and non-hateful hater reentry. Specifically, we experiment with instruction prompts, LLM finetuning, and LLM reinforcement learning (RL). Evaluation results show that our methods effectively steer the generation of counterspeech toward the desired outcomes. Our analyses, however, show that there are differences in the quality and style depending on the model.

Zexi Jia, Pengcheng Luo, Yijia Zhong et al.

Most evaluations of generative models rely on feature-distribution metrics such as FID, which operate on continuous recognition features that are explicitly trained to be invariant to appearance variations, and thus discard cues critical for perceptual quality. We instead evaluate models in the space of \emph{discrete} visual tokens, where modern 1D image tokenizers compactly encode both semantic and perceptual information and quality manifests as predictable token statistics. We introduce \emph{Codebook Histogram Distance} (CHD), a training-free distribution metric in token space, and \emph{Code Mixture Model Score} (CMMS), a no-reference quality metric learned from synthetic degradations of token sequences. To stress-test metrics under broad distribution shifts, we further propose \emph{VisForm}, a benchmark of 210K images spanning 62 visual forms and 12 generative models with expert annotations. Across AGIQA, HPDv2/3, and VisForm, our token-based metrics achieve state-of-the-art correlation with human judgments, and we will release all code and datasets to facilitate future research.

Xiaoying Song, Anirban Saha Anik, Dibakar Barua et al.

Health misinformation spreading online poses a significant threat to public health. Researchers have explored methods for automatically generating counterspeech to health misinformation as a mitigation strategy. Existing approaches often produce uniform responses, ignoring that the health literacy level of the audience could affect the accessibility and effectiveness of counterspeech. We propose a Controlled-Literacy framework using retrieval-augmented generation (RAG) with reinforcement learning (RL) to generate tailored counterspeech adapted to different health literacy levels. In particular, we retrieve knowledge aligned with specific health literacy levels, enabling accessible and factual information to support generation. We design a reward function incorporating subjective user preferences and objective readability-based rewards to optimize counterspeech to the target health literacy level. Experiment results show that Controlled-Literacy outperforms baselines by generating more accessible and user-preferred counterspeech. This research contributes to more equitable and impactful public health communication by improving the accessibility and comprehension of counterspeech to health misinformation

Pengcheng Luo, Yunyang Zhao, Bowen Zhang et al.

With the advent of 5G, the internet has entered a new video-centric era. From short-video platforms like TikTok to long-video platforms like Bilibili, online video services are reshaping user consumption habits. Adaptive Bitrate (ABR) control is widely recognized as a critical factor influencing Quality of Experience (QoE). Recent learning-based ABR methods have attracted increasing attention. However, most of them rely on limited network trace sets during training and overlook the wide-distribution characteristics of real-world network conditions, resulting in poor generalization in out-of-distribution (OOD) scenarios. To address this limitation, we propose SABR, a training framework that combines behavior cloning (BC) pretraining with reinforcement learning (RL) fine-tuning. We also introduce benchmarks, ABRBench-3G and ABRBench-4G+, which provide wide-coverage training traces and dedicated OOD test sets for assessing robustness to unseen network conditions. Experimental results demonstrate that SABR achieves the best average rank compared with Pensieve, Comyco, and NetLLM across the proposed benchmarks. These results indicate that SABR enables more stable learning across wide distributions and improves generalization to unseen network conditions.

Bowen Zhang, Pengcheng Luo, Genke Yang et al.

With the rise of artificial intelligence (AI), applying large language models (LLMs) to mathematical problem-solving has attracted increasing attention. Most existing approaches attempt to improve Operations Research (OR) optimization problem-solving through prompt engineering or fine-tuning strategies for LLMs. However, these methods are fundamentally constrained by the limited capabilities of non-reasoning LLMs. To overcome these limitations, we propose OR-LLM-Agent, an AI agent framework built on reasoning LLMs for automated OR problem solving. The framework decomposes the task into three sequential stages: mathematical modeling, code generation, and debugging. Each task is handled by a dedicated sub-agent, which enables more targeted reasoning. We also construct BWOR, an OR dataset for evaluating LLM performance on OR tasks. Our analysis shows that in the benchmarks NL4OPT, MAMO, and IndustryOR, reasoning LLMs sometimes underperform their non-reasoning counterparts within the same model family. In contrast, BWOR provides a more consistent and discriminative assessment of model capabilities. Experimental results demonstrate that OR-LLM-Agent utilizing DeepSeek-R1 in its framework outperforms advanced methods, including GPT-o3, Gemini 2.5 Pro, DeepSeek-R1, and ORLM, by at least 7\% in accuracy. These results demonstrate the effectiveness of task decomposition for OR problem solving.

Zhongliang Yang, Yongfeng Huang, Yiran Jiang et al.

Automatically extracting useful information from electronic medical records along with conducting disease diagnoses is a promising task for both clinical decision support(CDS) and neural language processing(NLP). Most of the existing systems are based on artificially constructed knowledge bases, and then auxiliary diagnosis is done by rule matching. In this study, we present a clinical intelligent decision approach based on Convolutional Neural Networks(CNN), which can automatically extract high-level semantic information of electronic medical records and then perform automatic diagnosis without artificial construction of rules or knowledge bases. We use collected 18,590 copies of the real-world clinical electronic medical records to train and test the proposed model. Experimental results show that the proposed model can achieve 98.67\% accuracy and 96.02\% recall, which strongly supports that using convolutional neural network to automatically learn high-level semantic features of electronic medical records and then conduct assist diagnosis is feasible and effective.