arXiv subjects

Hong Chen

Publications and source records attributed to Hong Chen.

At least 91 records · Page 5Linked to original sources

arXiv2025

Rethinking Key-frame-based Micro-expression Recognition: A Robust and Accurate Framework Against Key-frame Errors

Micro-expression recognition (MER) is a highly challenging task in affective computing. With the reduced-sized micro-expression (ME) input that contains key information based on key-frame indexes, key-frame-based methods have significantly improved the performance of MER. However, most of these methods focus on improving the performance with relatively accurate key-frame indexes, while ignoring the difficulty of obtaining accurate key-frame indexes and the objective existence of key-frame index errors, which impedes them from moving towards practical applications. In this paper, we propose CausalNet, a novel framework to achieve robust MER facing key-frame index errors while maintaining accurate recognition. To enhance robustness, CausalNet takes the representation of the entire ME sequence as the input. To address the information redundancy brought by the complete ME range input and maintain accurate recognition, first, the Causal Motion Position Learning Module (CMPLM) is proposed to help the model locate the muscle movement areas related to Action Units (AUs), thereby reducing the attention to other redundant areas. Second, the Causal Attention Block (CAB) is proposed to deeply learn the causal relationships between the muscle contraction and relaxation movements in MEs. Empirical experiments have demonstrated that on popular ME benchmarks, the CausalNet has achieved robust MER under different levels of key-frame index noise. Meanwhile, it has surpassed state-of-the-art (SOTA) methods on several standard MER benchmarks when using the provided annotated key-frames. Code is available at https://github.com/tony19980810/CausalNet.

cs.CV↗

arXiv2025

Monotonicity for generalized binomial coefficients and Jack positivity

Binomial formulas for Schur polynomials and Jack polynomials were studied by Lascoux in 1978, and Kaneko, Okounkov--Olshanski and Lassalle in the 1990s. We prove that the associated binomial coefficients are monotone and derive some symmetric function inequalities, in particular, a Schur positivity and Jack positivity result. These inequalities are similar to those studied by Newton, Muirhead, Gantmacher, Cuttler--Greene--Skandera, Sra and Khare--Tao.

math.CO↗

arXiv2025

How does Labeling Error Impact Contrastive Learning? A Perspective from Data Dimensionality Reduction

In recent years, contrastive learning has achieved state-of-the-art performance in the territory of self-supervised representation learning. Many previous works have attempted to provide the theoretical understanding underlying the success of contrastive learning. Almost all of them rely on a default assumption, i.e., the label consistency assumption, which may not hold in practice (the probability of failure is called labeling error) due to the strength and randomness of common augmentation strategies, such as random resized crop (RRC). This paper investigates the theoretical impact of labeling error on the downstream classification performance of contrastive learning. We first reveal several significant negative impacts of labeling error on downstream classification risk. To mitigate these impacts, data dimensionality reduction method (e.g., singular value decomposition, SVD) is applied on original data to reduce false positive samples, and establish both theoretical and empirical evaluations. Moreover, it is also found that SVD acts as a double-edged sword, which may lead to the deterioration of downstream classification accuracy due to the reduced connectivity of the augmentation graph. Based on the above observations, we give the augmentation suggestion that we should use some moderate embedding dimension (such as $512, 1024$ in our experiments), data inflation, weak augmentation, and SVD to ensure large graph connectivity and small labeling error to improve model performance.

stat.ML↗

arXiv2025

Clustering-Guided Multi-Layer Contrastive Representation Learning for Citrus Disease Classification

Citrus, as one of the most economically important fruit crops globally, suffers severe yield depressions due to various diseases. Accurate disease detection and classification serve as critical prerequisites for implementing targeted control measures. Recent advancements in artificial intelligence, particularly deep learning-based computer vision algorithms, have substantially decreased time and labor requirements while maintaining the accuracy of detection and classification. Nevertheless, these methods predominantly rely on massive, high-quality annotated training examples to attain promising performance. By introducing two key designs: contrasting with cluster centroids and a multi-layer contrastive training (MCT) paradigm, this paper proposes a novel clustering-guided self-supervised multi-layer contrastive representation learning (CMCRL) algorithm. The proposed method demonstrates several advantages over existing counterparts: (1) optimizing with massive unannotated samples; (2) effective adaptation to the symptom similarity across distinct citrus diseases; (3) hierarchical feature representation learning. The proposed method achieves state-of-the-art performance on the public citrus image set CDD, outperforming existing methods by 4.5\%-30.1\% accuracy. Remarkably, our method narrows the performance gap with fully supervised counterparts (all samples are labeled). Beyond classification accuracy, our method shows great performance on other evaluation metrics (F1 score, precision, and recall), highlighting the robustness against the class imbalance challenge.

cs.CV↗

arXiv2025

OmniSQL: Synthesizing High-quality Text-to-SQL Data at Scale

Text-to-SQL, the task of translating natural language questions into SQL queries, plays a crucial role in enabling non-experts to interact with databases. While recent advancements in large language models (LLMs) have significantly enhanced text-to-SQL performance, existing approaches face notable limitations in real-world text-to-SQL applications. Prompting-based methods often depend on closed-source LLMs, which are expensive, raise privacy concerns, and lack customization. Fine-tuning-based methods, on the other hand, suffer from poor generalizability due to the limited coverage of publicly available training data. To overcome these challenges, we propose a novel and scalable text-to-SQL data synthesis framework for automatically synthesizing large-scale, high-quality, and diverse datasets without extensive human intervention. Using this framework, we introduce SynSQL-2.5M, the first million-scale text-to-SQL dataset, containing 2.5 million samples spanning over 16,000 synthetic databases. Each sample includes a database, SQL query, natural language question, and chain-of-thought (CoT) solution. Leveraging SynSQL-2.5M, we develop OmniSQL, a powerful open-source text-to-SQL model available in three sizes: 7B, 14B, and 32B. Extensive evaluations across nine datasets demonstrate that OmniSQL achieves state-of-the-art performance, matching or surpassing leading closed-source and open-source LLMs, including GPT-4o and DeepSeek-V3, despite its smaller size. We release all code, datasets, and models to support further research.

cs.CL↗

arXiv2025

The Noisy Path from Source to Citation: Measuring How Scholars Engage with Past Research

Academic citations are widely used for evaluating research and tracing knowledge flows. Such uses typically rely on raw citation counts and neglect variability in citation types. In particular, citations can vary in their fidelity as original knowledge from cited studies may be paraphrased, summarized, or reinterpreted, possibly wrongly, leading to variation in how much information changes from cited to citing paper. In this study, we introduce a computational pipeline to quantify citation fidelity at scale. Using full texts of papers, the pipeline identifies citations in citing papers and the corresponding claims in cited papers, and applies supervised models to measure fidelity at the sentence level. Analyzing a large-scale multi-disciplinary dataset of approximately 13 million citation sentence pairs, we find that citation fidelity is higher when authors cite papers that are 1) more recent and intellectually close, 2) more accessible, and 3) the first author has a lower H-index and the author team is medium-sized. Using a quasi-experiment, we establish the "telephone effect" - when citing papers have low fidelity to the original claim, future papers that cite the citing paper and the original have lower fidelity to the original. Our work reveals systematic differences in citation fidelity, underscoring the limitations of analyses that rely on citation quantity alone and the potential for distortion of evidence.

cs.CL↗

arXiv2025

Dynamic Mixture of Curriculum LoRA Experts for Continual Multimodal Instruction Tuning

Continual multimodal instruction tuning is crucial for adapting Multimodal Large Language Models (MLLMs) to evolving tasks. However, most existing methods adopt a fixed architecture, struggling with adapting to new tasks due to static model capacity. We propose to evolve the architecture under parameter budgets for dynamic task adaptation, which remains unexplored and imposes two challenges: 1) task architecture conflict, where different tasks require varying layer-wise adaptations, and 2) modality imbalance, where different tasks rely unevenly on modalities, leading to unbalanced updates. To address these challenges, we propose a novel Dynamic Mixture of Curriculum LoRA Experts (D-MoLE) method, which automatically evolves MLLM's architecture with controlled parameter budgets to continually adapt to new tasks while retaining previously learned knowledge. Specifically, we propose a dynamic layer-wise expert allocator, which automatically allocates LoRA experts across layers to resolve architecture conflicts, and routes instructions layer-wisely to facilitate knowledge sharing among experts. Then, we propose a gradient-based inter-modal continual curriculum, which adjusts the update ratio of each module in MLLM based on the difficulty of each modality within the task to alleviate the modality imbalance problem. Extensive experiments show that D-MoLE significantly outperforms state-of-the-art baselines, achieving a 15% average improvement over the best baseline. To the best of our knowledge, this is the first study of continual learning for MLLMs from an architectural perspective.

cs.CV↗

arXiv2025

From Failures to Fixes: LLM-Driven Scenario Repair for Self-Evolving Autonomous Driving

Ensuring robust and generalizable autonomous driving requires not only broad scenario coverage but also efficient repair of failure cases, particularly those related to challenging and safety-critical scenarios. However, existing scenario generation and selection methods often lack adaptivity and semantic relevance, limiting their impact on performance improvement. In this paper, we propose \textbf{SERA}, an LLM-powered framework that enables autonomous driving systems to self-evolve by repairing failure cases through targeted scenario recommendation. By analyzing performance logs, SERA identifies failure patterns and dynamically retrieves semantically aligned scenarios from a structured bank. An LLM-based reflection mechanism further refines these recommendations to maximize relevance and diversity. The selected scenarios are used for few-shot fine-tuning, enabling targeted adaptation with minimal data. Experiments on the benchmark show that SERA consistently improves key metrics across multiple autonomous driving baselines, demonstrating its effectiveness and generalizability under safety-critical conditions.

cs.CV↗

arXiv2025

Personalized Clustering via Targeted Representation Learning

Clustering traditionally aims to reveal a natural grouping structure within unlabeled data. However, this structure may not always align with users' preferences. In this paper, we propose a personalized clustering method that explicitly performs targeted representation learning by interacting with users via modicum task information (e.g., $\textit{must-link}$ or $\textit{cannot-link}$ pairs) to guide the clustering direction. We query users with the most informative pairs, i.e., those pairs most hard to cluster and those most easy to miscluster, to facilitate the representation learning in terms of the clustering preference. Moreover, by exploiting attention mechanism, the targeted representation is learned and augmented. By leveraging the targeted representation and constrained contrastive loss as well, personalized clustering is obtained. Theoretically, we verify that the risk of personalized clustering is tightly bounded, guaranteeing that active queries to users do mitigate the clustering risk. Experimentally, extensive results show that our method performs well across different clustering tasks and datasets, even when only a limited number of queries are available.

cs.LG↗

arXiv2025

P$^2$ Law: Scaling Law for Post-Training After Model Pruning

Pruning has become a widely adopted technique for reducing the hardware requirements of large language models (LLMs). To recover model performance after pruning, post-training is commonly employed to mitigate the resulting performance degradation. While post-training benefits from larger datasets, once the dataset size is already substantial, increasing the training data provides only limited performance gains. To balance post-training cost and model performance, it is necessary to explore the optimal amount of post-training data.Through extensive experiments on the Llama-3 and Qwen-2.5 series models, pruned using various common pruning methods, we uncover the scaling \textbf{Law} for \textbf{P}ost-training after model \textbf{P}runing, referred to as the P$^2$ Law.This law identifies four key factors for predicting the pruned model's post-training loss: the model size before pruning, the number of post-training tokens, the pruning rate, and the model's loss before pruning. Moreover, P$^2$ Law can generalize to larger dataset sizes, larger model sizes, and higher pruning rates, offering valuable insights for the post-training of pruned LLMs.

cs.AI↗

arXiv2025

Chain-of-Thought for Autonomous Driving: A Comprehensive Survey and Future Prospects

The rapid evolution of large language models in natural language processing has substantially elevated their semantic understanding and logical reasoning capabilities. Such proficiencies have been leveraged in autonomous driving systems, contributing to significant improvements in system performance. Models such as OpenAI o1 and DeepSeek-R1, leverage Chain-of-Thought (CoT) reasoning, an advanced cognitive method that simulates human thinking processes, demonstrating remarkable reasoning capabilities in complex tasks. By structuring complex driving scenarios within a systematic reasoning framework, this approach has emerged as a prominent research focus in autonomous driving, substantially improving the system's ability to handle challenging cases. This paper investigates how CoT methods improve the reasoning abilities of autonomous driving models. Based on a comprehensive literature review, we present a systematic analysis of the motivations, methodologies, challenges, and future research directions of CoT in autonomous driving. Furthermore, we propose the insight of combining CoT with self-learning to facilitate self-evolution in driving systems. To ensure the relevance and timeliness of this study, we have compiled a dynamic repository of literature and open-source projects, diligently updated to incorporate forefront developments. The repository is publicly available at https://github.com/cuiyx1720/Awesome-CoT4AD.

cs.RO↗

arXiv2025

BehaveGPT: A Foundation Model for Large-scale User Behavior Modeling

In recent years, foundational models have revolutionized the fields of language and vision, demonstrating remarkable abilities in understanding and generating complex data; however, similar advances in user behavior modeling have been limited, largely due to the complexity of behavioral data and the challenges involved in capturing intricate temporal and contextual relationships in user activities. To address this, we propose BehaveGPT, a foundational model designed specifically for large-scale user behavior prediction. Leveraging transformer-based architecture and a novel pretraining paradigm, BehaveGPT is trained on vast user behavior datasets, allowing it to learn complex behavior patterns and support a range of downstream tasks, including next behavior prediction, long-term generation, and cross-domain adaptation. Our approach introduces the DRO-based pretraining paradigm tailored for user behavior data, which improves model generalization and transferability by equitably modeling both head and tail behaviors. Extensive experiments on real-world datasets demonstrate that BehaveGPT outperforms state-of-the-art baselines, achieving more than a 10% improvement in macro and weighted recall, showcasing its ability to effectively capture and predict user behavior. Furthermore, we measure the scaling law in the user behavior domain for the first time on the Honor dataset, providing insights into how model performance scales with increased data and parameter sizes.

cs.IR↗

arXiv2025

Large-area topological wireless power transfer

Topological wireless power transfer (WPT) technologies have attracted considerable interest due to their high transmission efficiency and robustness in coupled array configurations. However, conventional periodic and quasi-periodic topological chains exhibit limited adaptability in complex application scenarios, such as large-area simultaneous multi-load charging. In this work, we experimentally demonstrate a large-area topological defect state by constructing a gapless chain of uniformly coupled resonators at the interface of two topologically distinct Su-Schrieffer-Heeger (SSH) configurations. This topological defect state exhibits strong localization at multiple target sites, enabling efficient and concurrent wireless power delivery to spatially distributed loads. Furthermore, the unique wavefunction distribution enhances robustness against positional variations, ensuring stable energy transfer despite fluctuations in device placement. The proposed large-area topological framework offers fundamental insights into harnessing diverse topological states for advanced WPT applications, particularly in scenarios demanding spatial flexibility and multi-target energy delivery.

physics.app-ph↗

arXiv2025

Machine Learning Assisted Long-Range Wireless Power Transfer

Near-field magnetic resonance wireless power transfer (WPT) technology has garnered significant attention due to its broad application prospects in medical implants, electric vehicles, and robotics. Addressing the challenges faced by traditional WPT systems in frequency optimization and sensitivity to environmental disturbances, this study innovatively applies the gradient descent optimization algorithm to enhance a system with topological characteristics. Experimental results demonstrate that the machine learning-optimized Su-Schrieffer-Heeger (SSH)-like chain exhibits exceptional performance in transfer efficiency and system robustness. This achievement integrates non-Hermitian physics, topological physics, and machine learning, opening up new avenues and showcasing immense potential for the development of high-performance near-field wave functional devices.

physics.app-ph↗

arXiv2025

Unsupervised Learning for Class Distribution Mismatch

Class distribution mismatch (CDM) refers to the discrepancy between class distributions in training data and target tasks. Previous methods address this by designing classifiers to categorize classes known during training, while grouping unknown or new classes into an "other" category. However, they focus on semi-supervised scenarios and heavily rely on labeled data, limiting their applicability and performance. To address this, we propose Unsupervised Learning for Class Distribution Mismatch (UCDM), which constructs positive-negative pairs from unlabeled data for classifier training. Our approach randomly samples images and uses a diffusion model to add or erase semantic classes, synthesizing diverse training pairs. Additionally, we introduce a confidence-based labeling mechanism that iteratively assigns pseudo-labels to valuable real-world data and incorporates them into the training process. Extensive experiments on three datasets demonstrate UCDM's superiority over previous semi-supervised methods. Specifically, with a 60% mismatch proportion on Tiny-ImageNet dataset, our approach, without relying on labeled data, surpasses OpenMatch (with 40 labels per class) by 35.1%, 63.7%, and 72.5% in classifying known, unknown, and new classes.

cs.CV↗

arXiv2025

Non-Hermitian exceptional physics in RP^2 hyperbolic media

Conventional momentum space provides an orientable base space of a torus for topological classifications based on band theory. Here, we introduce a non-orientable momentum space isomorphic to the real projective plane RP^2 within the low-symmetry media. We show that the local band fluidity can be characterized by an expanded dihedral group with non-Abelian properties, while the global band fluidity offers a versatile platform to explore the evolution of non-Hermitian exceptional manifolds, including order-1, higher-order, hybrid exceptional manifolds, diabolic points and even bound states in the continuum. Furthermore, the non-orientable momentum space can pave the way for exploring the emergence of phenomena for exceptional manifolds.

physics.optics↗

arXiv2025

Thermally Induced Refractive Index Trimming of Visible-Light Silicon Nitride Waveguides Using Suspended Heaters

We demonstrate refractive index trimming of visible-light silicon nitride (SiN) waveguides using suspended heater structures. The thermal isolation of the suspended heaters enabled a semi-uniform temperature distribution with estimated temperatures of $\sim$350°C in the waveguides without reaching potentially damaging temperatures in the titanium nitride resistive heaters. The thermal isolation also enabled trimming temperatures to be reached with a moderate power dissipation of 30 to 40 mW. At a wavelength of 561 nm, modal effective index changes up to $-8.3 \times 10^{-3}$ were observed following thermal trimming, and the index changes were stable over an observation period of 97 days. The devices were fabricated as part of our visible-light integrated photonics platform on 200-mm diameter silicon wafers. The suspended heaters also functioned as efficient thermo-optic phase shifters with power dissipation for a $π$ phase shift of about $1.2-1.8$ mW. The trimming method was applied to set the bias points of thermo-optic Mach-Zehnder interferometer switches to reduce the bias power of five devices from $0.29-2.32$ mW to $0.1-0.16$ mW. Thermal trimming at a wavelength of 445 nm was also demonstrated. Through material analysis before and after thermal treatment, we hypothesize that index trimming of the silica (SiO$_2$) waveguide cladding may be a potential underlying mechanism. Additionally, via extrapolations of the measured trimming data, we estimate the thermal aging behavior of the SiN waveguides in the suspended heaters at lower (125 - 250°C) operating temperatures.

physics.optics↗

arXiv2025

Modular-Cam: Modular Dynamic Camera-view Video Generation with LLM

Text-to-Video generation, which utilizes the provided text prompt to generate high-quality videos, has drawn increasing attention and achieved great success due to the development of diffusion models recently. Existing methods mainly rely on a pre-trained text encoder to capture the semantic information and perform cross attention with the encoded text prompt to guide the generation of video. However, when it comes to complex prompts that contain dynamic scenes and multiple camera-view transformations, these methods can not decompose the overall information into separate scenes, as well as fail to smoothly change scenes based on the corresponding camera-views. To solve these problems, we propose a novel method, i.e., Modular-Cam. Specifically, to better understand a given complex prompt, we utilize a large language model to analyze user instructions and decouple them into multiple scenes together with transition actions. To generate a video containing dynamic scenes that match the given camera-views, we incorporate the widely-used temporal transformer into the diffusion model to ensure continuity within a single scene and propose CamOperator, a modular network based module that well controls the camera movements. Moreover, we propose AdaControlNet, which utilizes ControlNet to ensure consistency across scenes and adaptively adjusts the color tone of the generated video. Extensive qualitative and quantitative experiments prove our proposed Modular-Cam's strong capability of generating multi-scene videos together with its ability to achieve fine-grained control of camera movements. Generated results are available at https://modular-cam.github.io.

cs.CV↗