You Li

Spandan Madan, You Li, Mengmi Zhang et al.

We present a new perspective on bridging the generalization gap between biological and computer vision -- mimicking the human visual diet. While computer vision models rely on internet-scraped datasets, humans learn from limited 3D scenes under diverse real-world transformations with objects in natural context. Our results demonstrate that incorporating variations and contextual cues ubiquitous in the human visual training data (visual diet) significantly improves generalization to real-world transformations such as lighting, viewpoint, and material changes. This improvement also extends to generalizing from synthetic to real-world data -- all models trained with a human-like visual diet outperform specialized architectures by large margins when tested on natural image data. These experiments are enabled by our two key contributions: a novel dataset capturing scene context and diverse real-world transformations to mimic the human visual diet, and a transformer model tailored to leverage these aspects of the human visual diet. All data and source code can be accessed at https://github.com/Spandan-Madan/human_visual_diet.

Dewei Zhou, You Li, Fan Ma et al.

We introduce the Multi-Instance Generation (MIG) task, which focuses on generating multiple instances within a single image, each accurately placed at predefined positions with attributes such as category, color, and shape, strictly following user specifications. MIG faces three main challenges: avoiding attribute leakage between instances, supporting diverse instance descriptions, and maintaining consistency in iterative generation. To address attribute leakage, we propose the Multi-Instance Generation Controller (MIGC). MIGC generates multiple instances through a divide-and-conquer strategy, breaking down multi-instance shading into single-instance tasks with singular attributes, later integrated. To provide more types of instance descriptions, we developed MIGC++. MIGC++ allows attribute control through text \& images and position control through boxes \& masks. Lastly, we introduced the Consistent-MIG algorithm to enhance the iterative MIG ability of MIGC and MIGC++. This algorithm ensures consistency in unmodified regions during the addition, deletion, or modification of instances, and preserves the identity of instances when their attributes are changed. We introduce the COCO-MIG and Multimodal-MIG benchmarks to evaluate these methods. Extensive experiments on these benchmarks, along with the COCO-Position benchmark and DrawBench, demonstrate that our methods substantially outperform existing techniques, maintaining precise control over aspects including position, attribute, and quantity. Project page: https://github.com/limuloo/MIGC.

Ziyue Li, Yuchen Fang, You Li et al. · cmu, tsinghua

A timely detection of seizures for newborn infants with electroencephalogram (EEG) has been a common yet life-saving practice in the Neonatal Intensive Care Unit (NICU). However, it requires great human efforts for real-time monitoring, which calls for automated solutions to neonatal seizure detection. Moreover, the current automated methods focusing on adult epilepsy monitoring often fail due to (i) dynamic seizure onset location in human brains; (ii) different montages on neonates and (iii) huge distribution shift among different subjects. In this paper, we propose a deep learning framework, namely STATENet, to address the exclusive challenges with exquisite designs at the temporal, spatial and model levels. The experiments over the real-world large-scale neonatal EEG dataset illustrate that our framework achieves significantly better seizure detection performance.

Bei Lin, You Li, Ning Gui et al.

Unsupervised graph representation learning(GRL) aims to distill diverse graph information into task-agnostic embeddings without label supervision. Due to a lack of support from labels, recent representation learning methods usually adopt self-supervised learning, and embeddings are learned by solving a handcrafted auxiliary task(so-called pretext task). However, partially due to the irregular non-Euclidean data in graphs, the pretext tasks are generally designed under homophily assumptions and cornered in the low-frequency signals, which results in significant loss of other signals, especially high-frequency signals widespread in graphs with heterophily. Motivated by this limitation, we propose a multi-view perspective and the usage of diverse pretext tasks to capture different signals in graphs into embeddings. A novel framework, denoted as Multi-view Graph Encoder(MVGE), is proposed, and a set of key designs are identified. More specifically, a set of new pretext tasks are designed to encode different types of signals, and a straightforward operation is propxwosed to maintain both the commodity and personalization in both the attribute and the structural levels. Extensive experiments on synthetic and real-world network datasets show that the node representations learned with MVGE achieve significant performance improvements in three different downstream tasks, especially on graphs with heterophily. Source code is available at \url{https://github.com/G-AILab/MVGE}.

You Li, Julien Moreau, Javier Ibanez-Guzman

Autonomous vehicles rely on perception systems to understand their surroundings for further navigation missions. Cameras are essential for perception systems due to the advantages of object detection and recognition provided by modern computer vision algorithms, comparing to other sensors, such as LiDARs and radars. However, limited by its inherent imaging principle, a standard RGB camera may perform poorly in a variety of adverse scenarios, including but not limited to: low illumination, high contrast, bad weather such as fog/rain/snow, etc. Meanwhile, estimating the 3D information from the 2D image detection is generally more difficult when compared to LiDARs or radars. Several new sensing technologies have emerged in recent years to address the limitations of conventional RGB cameras. In this paper, we review the principles of four novel image sensors: infrared cameras, range-gated cameras, polarization cameras, and event cameras. Their comparative advantages, existing or potential applications, and corresponding data processing algorithms are all presented in a systematic manner. We expect that this study will assist practitioners in the autonomous driving society with new perspectives and insights.

Parantak Singh, You Li, Ankur Sikarwar et al.

Curriculum design is a fundamental component of education. For example, when we learn mathematics at school, we build upon our knowledge of addition to learn multiplication. These and other concepts must be mastered before our first algebra lesson, which also reinforces our addition and multiplication skills. Designing a curriculum for teaching either a human or a machine shares the underlying goal of maximizing knowledge transfer from earlier to later tasks, while also minimizing forgetting of learned tasks. Prior research on curriculum design for image classification focuses on the ordering of training examples during a single offline task. Here, we investigate the effect of the order in which multiple distinct tasks are learned in a sequence. We focus on the online class-incremental continual learning setting, where algorithms or humans must learn image classes one at a time during a single pass through a dataset. We find that curriculum consistently influences learning outcomes for humans and for multiple continual machine learning algorithms across several benchmark datasets. We introduce a novel-object recognition dataset for human curriculum learning experiments and observe that curricula that are effective for humans are highly correlated with those that are effective for machines. As an initial step towards automated curriculum design for online class-incremental learning, we propose a novel algorithm, dubbed Curriculum Designer (CD), that designs and ranks curricula based on inter-class feature similarities. We find significant overlap between curricula that are empirically highly effective and those that are highly ranked by our CD. Our study establishes a framework for further research on teaching humans and machines to learn continuously using optimized curricula.

Xingchen Zhou, Ying He, F. Richard Yu et al.

The emergence of Neural Radiance Fields (NeRF) has promoted the development of synthesized high-fidelity views of the intricate real world. However, it is still a very demanding task to repaint the content in NeRF. In this paper, we propose a novel framework that can take RGB images as input and alter the 3D content in neural scenes. Our work leverages existing diffusion models to guide changes in the designated 3D content. Specifically, we semantically select the target object and a pre-trained diffusion model will guide the NeRF model to generate new 3D objects, which can improve the editability, diversity, and application range of NeRF. Experiment results show that our algorithm is effective for editing 3D objects in NeRF under different text prompts, including editing appearance, shape, and more. We validate our method on both real-world datasets and synthetic-world datasets for these editing tasks. Please visit https://starstesla.github.io/repaintnerf for a better view of our results.

Fanhu Zeng, Zhicong Luo, Zefan Wang et al.

Visual reasoning through reinforcement learning with verifiable rewards (RLVR) has achieved remarkable progress. However, when dealing with multi-source inputs, existing approaches tend to treat them as a mere accumulation of information, lacking explicit mechanisms to distinguish whether integrating additional sources yields information gain or introduces interference. Therefore, they struggle to effectively model dynamic interaction when integrating multiple sources, particularly when they differ significantly in physical properties and semantics, e.g., infrared and depth, leading to inferior performance to mono-source reasoning when a certain source holds the dominant signal. To address this issue, we propose MARS, a novel mono-anchored multi-source reasoning framework that models each visual modality as an independent information source. Specifically, by treating mono-source rewards as dynamic anchors, our method explicitly incorporates the information gain introduced by multi-source fusion into advantage normalization and adaptively emphasizes mutual promotion between sources while suppressing potential noise or conflicts during RLVR. From theoretical analysis, our method effectively quantifies information gain introduced by multi-source integration in gradient estimation, enabling consistent modality regulation. Empirical results also show impressive 3.2% and 4.9% performance gains on GRPO and DAPO across diverse datasets, confirming effectiveness of our method.

Junbo Cui, Bokai Xu, Chongyi Wang et al.

Recent progress in multimodal large language models (MLLMs) has brought AI capabilities from static offline data processing to real-time streaming interaction, yet they still remain far from human-level multimodal interaction. The key bottlenecks are no longer modality coverage or latency alone, but the interaction paradigm itself. First, perception and response are still separated into alternating phases, preventing models from incorporating new inputs for timely adjustment during generation. Second, most current models remain reactive, responding only to explicit user requests instead of acting proactively in the evolving multimodal environment. We present MiniCPM-o 4.5, our latest effort towards human-like multimodal interaction, which mitigates these gaps by real-time full-duplex omni-modal interaction. It can see, listen, and speak simultaneously in real-time, while also exhibiting proactive behaviors such as issuing reminders or comments based on its continuous understanding of the live scene. The key technique behind MiniCPM-o 4.5 is Omni-Flow, a unified streaming framework that aligns omni-modal inputs and outputs along a shared temporal axis. This formulation converts conventional turn-based interaction into a full-duplex, time-aligned process, enabling simultaneous perception and response and allowing proactive behavior to arise within the same framework. With a total of 9B parameters, MiniCPM-o 4.5 approaches Gemini 2.5 Flash in vision-language capabilities, delivering state-of-the-art open-source performance at its scale. It also surpasses Qwen3-Omni-30B-A3B in omni-modal understanding and delivers better speech generation, with significantly higher computation efficiency. Driven by its efficient architecture design and inference optimization, the model can perform real-time full-duplex omni-modal interaction on edge devices with less than 12GB RAM cost.

Shuyu Kong, You Li, Hai Zhou

Federated Learning is an emerging distributed collaborative learning paradigm used by many of applications nowadays. The effectiveness of federated learning relies on clients' collective efforts and their willingness to contribute local data. However, due to privacy concerns and the costs of data collection and model training, clients may not always contribute all the data they possess, which would negatively affect the performance of the global model. This paper presents an incentive mechanism that encourages clients to contribute as much data as they can obtain. Unlike previous incentive mechanisms, our approach does not monetize data. Instead, we implicitly use model performance as a reward, i.e., significant contributors are paid off with better models. We theoretically prove that clients will use as much data as they can possibly possess to participate in federated learning under certain conditions with our incentive mechanism

You Li, Bei Lin, Binli Luo et al.

Unsupervised graph representation learning aims to distill various graph information into a downstream task-agnostic dense vector embedding. However, existing graph representation learning approaches are designed mainly under the node homophily assumption: connected nodes tend to have similar labels and optimize performance on node-centric downstream tasks. Their design is apparently against the task-agnostic principle and generally suffers poor performance in tasks, e.g., edge classification, that demands feature signals beyond the node-view and homophily assumption. To condense different feature signals into the embeddings, this paper proposes PairE, a novel unsupervised graph embedding method using two paired nodes as the basic unit of embedding to retain the high-frequency signals between nodes to support node-related and edge-related tasks. Accordingly, a multi-self-supervised autoencoder is designed to fulfill two pretext tasks: one retains the high-frequency signal better, and another enhances the representation of commonality. Our extensive experiments on a diversity of benchmark datasets clearly show that PairE outperforms the unsupervised state-of-the-art baselines, with up to 101.1\% relative improvement on the edge classification tasks that rely on both the high and low-frequency signals in the pair and up to 82.5\% relative performance gain on the node classification tasks.

Jing Tao, You Li, Banglei Guan et al.

Under challenging light conditions, captured images often suffer from various degradations, leading to a decline in the performance of vision-based applications. Although numerous methods have been proposed to enhance image quality, they either significantly amplify inherent noise or are only effective under specific illumination conditions. To address these issues, we propose a novel framework for simultaneous enhancement and noise suppression under complex illumination conditions. Firstly, a gradient-domain weighted guided filter (GDWGIF) is employed to accurately estimate illumination and improve image quality. Next, the Retinex model is applied to decompose the captured image into separate illumination and reflection layers. These layers undergo parallel processing, with the illumination layer being corrected to optimize lighting conditions and the reflection layer enhanced to improve image quality. Finally, the dynamic range of the image is optimized through multi-exposure fusion and a linear stretching strategy. The proposed method is evaluated on real-world datasets obtained from practical applications. Experimental results demonstrate that our proposed method achieves better performance compared to state-of-the-art methods in both contrast enhancement and noise suppression.

You Li, Dewei Zhou, Fan Ma et al.

Recent Video-to-Audio (V2A) methods have achieved remarkable progress, enabling the synthesis of realistic, high-quality audio. However, they struggle with fine-grained temporal control in multi-event scenarios or when visual cues are insufficient, such as small regions, off-screen sounds, or occluded or partially visible objects. In this paper, we propose FoleyDirector, a framework that, for the first time, enables precise temporal guidance in DiT-based V2A generation while preserving the base model's audio quality and allowing seamless switching between V2A generation and temporally controlled synthesis. FoleyDirector introduces Structured Temporal Scripts (STS), a set of captions corresponding to short temporal segments, to provide richer temporal information. These features are integrated via the Script-Guided Temporal Fusion Module, which employs Temporal Script Attention to fuse STS features coherently. To handle complex multi-event scenarios, we further propose Bi-Frame Sound Synthesis, enabling parallel in-frame and out-of-frame audio generation and improving controllability. To support training and evaluation, we construct the DirectorSound dataset and introduce VGGSoundDirector and DirectorBench. Experiments demonstrate that FoleyDirector substantially enhances temporal controllability while maintaining high audio fidelity, empowering users to act as Foley directors and advancing V2A toward more expressive and controllable generation.

You Li, Heyu Huang, Chi Chen et al.

The recent advancement of Multimodal Large Language Models (MLLMs) has significantly improved their fine-grained perception of single images and general comprehension across multiple images. However, existing MLLMs still face challenges in achieving precise grounding in complex multi-image scenarios. To address this, we first explore a Chain-of-Thought (CoT) framework that integrates single-image grounding with multi-image comprehension. While partially effective, it remains unstable and struggles to capture abstract visual information due to its non-end-to-end nature. Therefore, we introduce Migician, the first multi-image grounding model capable of performing free-form and accurate grounding across multiple images. To support this, we present the MGrounding-630k dataset, which comprises data for several multi-image grounding tasks derived from existing datasets, along with newly generated free-form grounding instruction-following data. Furthermore, we propose MIG-Bench, a comprehensive benchmark specifically designed for evaluating multi-image grounding capabilities. Experimental results demonstrate that our model achieves significantly superior multi-image grounding capabilities, outperforming the best existing MLLMs by 24.94% and even surpassing much larger 70B models. Our code, model, dataset, and benchmark are fully open-sourced at https://migician-vg.github.io/.

Delong Wu, Hao Zhu, Qi Zhang et al.

Implicit Neural Representation (INR) has become a popular method for representing visual signals (e.g., 2D images and 3D scenes), demonstrating promising results in various downstream applications. Given its potential as a medium for visual signals, exploring the development of a neural blending method that utilizes INRs is a natural progression. Neural blending involves merging two INRs to create a new INR that encapsulates information from both original representations. A direct approach involves applying traditional image editing methods to the INR rendering process. However, this method often results in blending distortions, artifacts, and color shifts, primarily due to the discretization of the underlying pixel grid and the introduction of boundary conditions for solving variational problems. To tackle this issue, we introduce the Neural Poisson Solver, a plug-and-play and universally applicable framework across different signal dimensions for blending visual signals represented by INRs. Our Neural Poisson Solver offers a variational problem-solving approach based on the continuous Poisson equation, demonstrating exceptional performance across various domains. Specifically, we propose a gradient-guided neural solver to represent the solution process of the variational problem, refining the target signal to achieve natural blending results. We also develop a Poisson equation-based loss and optimization scheme to train our solver, ensuring it effectively blends the input INR scenes while preserving their inherent structure and semantic content. The lack of dependence on additional prior knowledge makes our method easily adaptable to various task categories, highlighting its versatility. Comprehensive experimental results validate the robustness of our approach across multiple dimensions and blending tasks.

Xinyue Lou, You Li, Jinan Xu et al.

The rapid development of Multimodal Large Reasoning Models (MLRMs) has demonstrated broad application potential, yet their safety and reliability remain critical concerns that require systematic exploration. To address this gap, we conduct a comprehensive and systematic safety evaluation of 11 MLRMs across 5 benchmarks and unveil prevalent safety degradation phenomena in most advanced models. Moreover, our analysis reveals distinct safety patterns across different benchmarks: significant safety degradation is observed across jailbreak robustness benchmarks, whereas safety-awareness benchmarks demonstrate less pronounced degradation. In particular, the long thought process in some scenarios even enhances safety performance. Therefore, it is a potential approach to address safety issues in MLRMs by leveraging the intrinsic reasoning capabilities of the model to detect unsafe intent. To operationalize this insight, we construct a multimodal tuning dataset that incorporates a safety-oriented thought process. Experimental results from fine-tuning existing MLRMs with this dataset effectively enhances the safety on both jailbreak robustness and safety-awareness benchmarks. This study provides a new perspective for developing safe MLRMs. Our dataset is available at https://github.com/xinyuelou/Think-in-Safety.

Long Yuan, Fengran Mo, Kaiyu Huang et al.

The rapid advancement of multimodal large language models (LLMs) has opened new frontiers in artificial intelligence, enabling the integration of diverse large-scale data types such as text, images, and spatial information. In this paper, we explore the potential of multimodal LLMs (MLLM) for geospatial artificial intelligence (GeoAI), a field that leverages spatial data to address challenges in domains including Geospatial Semantics, Health Geography, Urban Geography, Urban Perception, and Remote Sensing. We propose a MLLM (OmniGeo) tailored to geospatial applications, capable of processing and analyzing heterogeneous data sources, including satellite imagery, geospatial metadata, and textual descriptions. By combining the strengths of natural language understanding and spatial reasoning, our model enhances the ability of instruction following and the accuracy of GeoAI systems. Results demonstrate that our model outperforms task-specific models and existing LLMs on diverse geospatial tasks, effectively addressing the multimodality nature while achieving competitive results on the zero-shot geospatial tasks. Our code will be released after publication.

You Li, Chi Chen, Yanghao Li et al.

Latent visual reasoning aims to mimic human's imagination process by meditating through hidden states of Multimodal Large Language Models. While recognized as a promising paradigm for visual reasoning, the underlying mechanisms driving its effectiveness remain unclear. Motivated to demystify the true source of its efficacy, we investigate the validity of latent reasoning using Causal Mediation Analysis. We model the process as a causal chain: the input as the treatment, the latent tokens as the mediator, and the final answer as the outcome. Our findings uncover two critical disconnections: (a) Input-Latent Disconnect: dramatic perturbations on the input result in negligible changes to the latent tokens, suggesting that latent tokens do not effectively attend to the input sequence. (b) Latent-Answer Disconnect: perturbations on the latent tokens yield minimal impact on the final answer, indicating the limited causal effect latent tokens imposing on the outcome. Furthermore, extensive probing analysis reveals that latent tokens encode limited visual information and exhibit high similarity. Consequently, we challenge the necessity of latent reasoning and propose a straightforward alternative named CapImagine, which teaches the model to explicitly imagine using text. Experiments on vision-centric benchmarks show that CapImagine significantly outperforms complex latent-space baselines, highlighting the superior potential of visual reasoning through explicit imagination.

Runhao Liu, You Li, Zhengyang Cheng et al.

Long-horizon agricultural planning requires optimizing crop allocation under complex spatial heterogeneity, temporal agronomic dependencies, and multi-source environmental uncertainty. Existing approaches often either address crop interactions, such as legume-cereal complementarity, only implicitly or rely on static deterministic formulations that fail to ensure resilience against market and climate volatility.To address these challenges, we propose a Multi-Layer Robust Crop Planning Framework (MLRCPF) that integrates spatial reasoning, temporal dynamics, and robust optimization. Specifically, we formalize crop-to-crop relationships through a structured interaction matrix embedded within the state-transition logic, and employ a distributionally robust optimization layer to mitigate worst-case risks defined by a data-driven ambiguity set. Evaluations on a real-world high-mix farming dataset from North China demonstrate the effectiveness of the proposed approach. The framework autonomously generates sustainable checkerboard rotation patterns that restore soil fertility, significantly increasing the legume planting ratio compared to deterministic baselines. Economically, it successfully resolves the trade-off between optimality and stability. These results highlight the importance of explicitly encoding domain-specific structural priors into optimization models for resilient decision-making in complex agricultural systems.

You Li, Jinhui Yin, Yuming Lin

Pretrained language models are expected to effectively map input text to a set of vectors while preserving the inherent relationships within the text. Consequently, designing a white-box model to compute metrics that reflect the presence of specific internal relations in these vectors has become a common approach for post-hoc interpretability analysis of pretrained language models. However, achieving interpretability in white-box models and ensuring the rigor of metric computation becomes challenging when the source model lacks inherent interpretability. Therefore, in this paper, we discuss striking a balance in this trade-off and propose a novel line to constructing metrics for understanding the mechanisms of pretrained language models. We have specifically designed a family of metrics along this line of investigation, and the model used to compute these metrics is referred to as the tree topological probe. We conducted measurements on BERT-large by using these metrics. Based on the experimental results, we propose a speculation regarding the working mechanism of BERT-like pretrained language models, as well as a strategy for enhancing fine-tuning performance by leveraging the topological probe to improve specific submodules.

You Li, Guannan Zhao, Yuhao Ju et al.

We introduce LLA, an effective intellectual property (IP) protection scheme for generative AI models. LLA leverages the synergy between hardware and software to defend against various supply chain threats, including model theft, model corruption, and information leakage. On the software side, it embeds key bits into neurons that can trigger outliers to degrade performance and applies invariance transformations to obscure the key values. On the hardware side, it integrates a lightweight locking module into the AI accelerator while maintaining compatibility with various dataflow patterns and toolchains. An accelerator with a pre-stored secret key acts as a license to access the model services provided by the IP owner. The evaluation results show that LLA can withstand a broad range of oracle-guided key optimization attacks, while incurring a minimal computational overhead of less than 0.1% for 7,168 key bits.

Qishan Wang, Shuyong Gao, Junjie Hu et al.

Multi-class Unsupervised Anomaly Detection algorithms (MUAD) are receiving increasing attention due to their relatively low deployment costs and improved training efficiency. However, the real-world effectiveness of MUAD methods is questioned due to limitations in current Industrial Anomaly Detection (IAD) datasets. These datasets contain numerous classes that are unlikely to be produced by the same factory and fail to cover multiple structures or appearances. Additionally, the defects do not reflect real-world characteristics. Therefore, we introduce the Heterogeneous Same-Sort Industrial Anomaly Detection (HSS-IAD) dataset, which contains 8,580 images of metallic-like industrial parts and precise anomaly annotations. These parts exhibit variations in structure and appearance, with subtle defects that closely resemble the base materials. We also provide foreground images for synthetic anomaly generation. Finally, we evaluate popular IAD methods on this dataset under multi-class and class-separated settings, demonstrating its potential to bridge the gap between existing datasets and real factory conditions. The dataset is available at https://github.com/Qiqigeww/HSS-IAD-Dataset.

Weiran Li, Zhenbo Li, Fei Li et al.

Fish tracking is a key technology for obtaining movement trajectories and identifying abnormal behavior. However, it faces considerable challenges, including occlusion, multi-scale tracking, and fish deformation. Notably, extant reviews have focused more on behavioral analysis rather than providing a comprehensive overview of computer vision-based fish tracking approaches. This paper presents a comprehensive review of the advancements of fish tracking technologies over the past seven years (2017-2023). It explores diverse fish tracking techniques with an emphasis on fundamental localization and tracking methods. Auxiliary plugins commonly integrated into fish tracking systems, such as underwater image enhancement and re-identification, are also examined. Additionally, this paper summarizes open-source datasets, evaluation metrics, challenges, and applications in fish tracking research. Finally, a comprehensive discussion offers insights and future directions for vision-based fish tracking techniques. We hope that our work could provide a partial reference in the development of fish tracking algorithms.

Dewei Zhou, You Li, Fan Ma et al.

We present a Multi-Instance Generation (MIG) task, simultaneously generating multiple instances with diverse controls in one image. Given a set of predefined coordinates and their corresponding descriptions, the task is to ensure that generated instances are accurately at the designated locations and that all instances' attributes adhere to their corresponding description. This broadens the scope of current research on Single-instance generation, elevating it to a more versatile and practical dimension. Inspired by the idea of divide and conquer, we introduce an innovative approach named Multi-Instance Generation Controller (MIGC) to address the challenges of the MIG task. Initially, we break down the MIG task into several subtasks, each involving the shading of a single instance. To ensure precise shading for each instance, we introduce an instance enhancement attention mechanism. Lastly, we aggregate all the shaded instances to provide the necessary information for accurately generating multiple instances in stable diffusion (SD). To evaluate how well generation models perform on the MIG task, we provide a COCO-MIG benchmark along with an evaluation pipeline. Extensive experiments were conducted on the proposed COCO-MIG benchmark, as well as on various commonly used benchmarks. The evaluation results illustrate the exceptional control capabilities of our model in terms of quantity, position, attribute, and interaction. Code and demos will be released at https://migcproject.github.io/.

Kaiyu Huang, Fengran Mo, Xinyu Zhang et al. · tsinghua

The rapid development of Large Language Models (LLMs) demonstrates remarkable multilingual capabilities in natural language processing, attracting global attention in both academia and industry. To mitigate potential discrimination and enhance the overall usability and accessibility for diverse language user groups, it is important for the development of language-fair technology. Despite the breakthroughs of LLMs, the investigation into the multilingual scenario remains insufficient, where a comprehensive survey to summarize recent approaches, developments, limitations, and potential solutions is desirable. To this end, we provide a survey with multiple perspectives on the utilization of LLMs in the multilingual scenario. We first rethink the transitions between previous and current research on pre-trained language models. Then we introduce several perspectives on the multilingualism of LLMs, including training and inference methods, information retrieval, model security, multi-domain with language culture, and usage of datasets. We also discuss the major challenges that arise in these aspects, along with possible solutions. Besides, we highlight future research directions that aim at further enhancing LLMs with multilingualism. The survey aims to help the research community address multilingual problems and provide a comprehensive understanding of the core concepts, key techniques, and latest developments in multilingual natural language processing based on LLMs.

Mengwen Ye, Yingzi Huangfu, You Li et al.

Fundus image classification is crucial in the computer aided diagnosis tasks, but label noise significantly impairs the performance of deep neural networks. To address this challenge, we propose a robust framework, Self-Supervised Pre-training with Robust Adaptive Credal Loss (SSP-RACL), for handling label noise in fundus image datasets. First, we use Masked Autoencoders (MAE) for pre-training to extract features, unaffected by label noise. Subsequently, RACL employ a superset learning framework, setting confidence thresholds and adaptive label relaxation parameter to construct possibility distributions and provide more reliable ground-truth estimates, thus effectively suppressing the memorization effect. Additionally, we introduce clinical knowledge-based asymmetric noise generation to simulate real-world noisy fundus image datasets. Experimental results demonstrate that our proposed method outperforms existing approaches in handling label noise, showing superior performance.

Dewei Zhou, You Li, Zongxin Yang et al.

We introduce region-specific image refinement as a dedicated problem setting: given an input image and a user-specified region (e.g., a scribble mask or a bounding box), the goal is to restore fine-grained details while keeping all non-edited pixels strictly unchanged. Despite rapid progress in image generation, modern models still frequently suffer from local detail collapse (e.g., distorted text, logos, and thin structures). Existing instruction-driven editing models emphasize coarse-grained semantic edits and often either overlook subtle local defects or inadvertently change the background, especially when the region of interest occupies only a small portion of a fixed-resolution input. We present RefineAnything, a multimodal diffusion-based refinement model that supports both reference-based and reference-free refinement. Building on a counter-intuitive observation that crop-and-resize can substantially improve local reconstruction under a fixed VAE input resolution, we propose Focus-and-Refine, a region-focused refinement-and-paste-back strategy that improves refinement effectiveness and efficiency by reallocating the resolution budget to the target region, while a blended-mask paste-back guarantees strict background preservation. We further introduce a boundary-aware Boundary Consistency Loss to reduce seam artifacts and improve paste-back naturalness. To support this new setting, we construct Refine-30K (20K reference-based and 10K reference-free samples) and introduce RefineEval, a benchmark that evaluates both edited-region fidelity and background consistency. On RefineEval, RefineAnything achieves strong improvements over competitive baselines and near-perfect background preservation, establishing a practical solution for high-precision local refinement. Project Page: https://limuloo.github.io/RefineAnything/.

You Li, Fan Ma, Yi Yang

The Zero-shot Composed Image Retrieval (ZSCIR) requires retrieving images that match the query image and the relative captions. Current methods focus on projecting the query image into the text feature space, subsequently combining them with features of query texts for retrieval. However, retrieving images only with the text features cannot guarantee detailed alignment due to the natural gap between images and text. In this paper, we introduce Imagined Proxy for CIR (IP-CIR), a training-free method that creates a proxy image aligned with the query image and text description, enhancing query representation in the retrieval process. We first leverage the large language model's generalization capability to generate an image layout, and then apply both the query text and image for conditional generation. The robust query features are enhanced by merging the proxy image, query image, and text semantic perturbation. Our newly proposed balancing metric integrates text-based and proxy retrieval similarities, allowing for more accurate retrieval of the target image while incorporating image-side information into the process. Experiments on three public datasets demonstrate that our method significantly improves retrieval performances. We achieve state-of-the-art (SOTA) results on the CIRR dataset with a Recall@K of 70.07 at K=10. Additionally, we achieved an improvement in Recall@10 on the FashionIQ dataset, rising from 45.11 to 45.74, and improved the baseline performance in CIRCO with a mAPK@10 score, increasing from 32.24 to 34.26.

Xiang Li, You Li, Yazhou Zhang

EEG-based emotion recognition is hampered by profound dataset heterogeneity (channel/subject variability), hindering generalizable models. Existing approaches struggle to transfer knowledge effectively. We propose 'One Model for All', a universal pre-training framework for EEG analysis across disparate datasets. Our paradigm decouples learning into two stages: (1) Univariate pre-training via self-supervised contrastive learning on individual channels, enabled by a Unified Channel Schema (UCS) that leverages the channel union (e.g., SEED-62ch, DEAP-32ch); (2) Multivariate fine-tuning with a novel 'ART' (Adaptive Resampling Transformer) and 'GAT' (Graph Attention Network) architecture to capture complex spatio-temporal dependencies. Experiments show universal pre-training is an essential stabilizer, preventing collapse on SEED (vs. scratch) and yielding substantial gains on DEAP (+7.65%) and DREAMER (+3.55%). Our framework achieves new SOTA performance on all within-subject benchmarks: SEED (99.27%), DEAP (93.69%), and DREAMER (93.93%). We also show SOTA cross-dataset transfer, achieving 94.08% (intersection) and 93.05% (UCS) on the unseen DREAMER dataset, with the former surpassing the within-domain pre-training benchmark. Ablation studies validate our architecture: the GAT module is critical, yielding a +22.19% gain over GCN on the high-noise DEAP dataset, and its removal causes a catastrophic -16.44% performance drop. This work paves the way for more universal, scalable, and effective pre-trained models for diverse EEG analysis tasks.

Zongjie Li, Wenying Qiu, Pingchuan Ma et al.

Recent years have witnessed the rapid development of large language models (LLMs) in various domains. To better serve the large number of Chinese users, many commercial vendors in China have adopted localization strategies, training and providing local LLMs specifically customized for Chinese users. Furthermore, looking ahead, one of the key future applications of LLMs will be practical deployment in industrial production by enterprises and users in those sectors. However, the accuracy and robustness of LLMs in industrial scenarios have not been well studied. In this paper, we present a comprehensive empirical study on the accuracy and robustness of LLMs in the context of the Chinese industrial production area. We manually collected 1,200 domain-specific problems from 8 different industrial sectors to evaluate LLM accuracy. Furthermore, we designed a metamorphic testing framework containing four industrial-specific stability categories with eight abilities, totaling 13,631 questions with variants to evaluate LLM robustness. In total, we evaluated 9 different LLMs developed by Chinese vendors, as well as four different LLMs developed by global vendors. Our major findings include: (1) Current LLMs exhibit low accuracy in Chinese industrial contexts, with all LLMs scoring less than 0.6. (2) The robustness scores vary across industrial sectors, and local LLMs overall perform worse than global ones. (3) LLM robustness differs significantly across abilities. Global LLMs are more robust under logical-related variants, while advanced local LLMs perform better on problems related to understanding Chinese industrial terminology. Our study results provide valuable guidance for understanding and promoting the industrial domain capabilities of LLMs from both development and industrial enterprise perspectives. The results further motivate possible research directions and tooling support.

You Li, Fan Ma, Yi Yang

Diffusion models have recently been employed to generate high-quality images, reducing the need for manual data collection and improving model generalization in tasks such as object detection, instance segmentation, and image perception. However, the synthetic framework is usually designed with meticulous human effort for each task due to various requirements on image layout, content, and annotation formats, restricting the application of synthetic data on more general scenarios. In this paper, we propose AnySynth, a unified framework integrating adaptable, comprehensive, and highly controllable components capable of generating an arbitrary type of synthetic data given diverse requirements. Specifically, the Task-Specific Layout Generation Module is first introduced to produce reasonable layouts for different tasks by leveraging the generation ability of large language models and layout priors of real-world images. A Uni-Controlled Image Generation Module is then developed to create high-quality synthetic images that are controllable and based on the generated layouts. In addition, user specific reference images, and style images can be incorporated into the generation to task requirements. Finally, the Task-Oriented Annotation Module offers precise and detailed annotations for the generated images across different tasks. We have validated our framework's performance across various tasks, including Few-shot Object Detection, Cross-domain Object Detection, Zero-shot Composed Image Retrieval, and Multi-modal Image Perception and Grounding. The specific data synthesized by our framework significantly improves model performance in these tasks, demonstrating the generality and effectiveness of our framework.

Jiawen Yu, Jieji Ren, Yang Chang et al.

Anomaly detection and localization in automated industrial manufacturing can significantly enhance production efficiency and product quality. Existing methods are capable of detecting surface defects in pre-defined or controlled imaging environments. However, accurately detecting workpiece defects in complex and unstructured industrial environments with varying views, poses and illumination remains challenging. We propose a novel anomaly detection and localization method specifically designed to handle inputs with perturbative patterns. Our approach introduces a new framework based on a collaborative distillation heterogeneous teacher network (HetNet), an adaptive local-global feature fusion module, and a local multivariate Gaussian noise generation module. HetNet can learn to model the complex feature distribution of normal patterns using limited information about local disruptive changes. We conducted extensive experiments on mainstream benchmarks. HetNet demonstrates superior performance with approximately 10% improvement across all evaluation metrics on MSC-AD under industrial conditions, while achieving state-of-the-art results on other datasets, validating its resilience to environmental fluctuations and its capability to enhance the reliability of industrial anomaly detection systems across diverse scenarios. Tests in real-world environments further confirm that HetNet can be effectively integrated into production lines to achieve robust and real-time anomaly detection. Codes, images and videos are published on the project website at: https://zihuatanejoyu.github.io/HetNet/

Ning Cheng, You Li, Jing Gao et al.

Tactility provides crucial support and enhancement for the perception and interaction capabilities of both humans and robots. Nevertheless, the multimodal research related to touch primarily focuses on visual and tactile modalities, with limited exploration in the domain of language. Beyond vocabulary, sentence-level descriptions contain richer semantics. Based on this, we construct a touch-language-vision dataset named TLV (Touch-Language-Vision) by human-machine cascade collaboration, featuring sentence-level descriptions for multimode alignment. The new dataset is used to fine-tune our proposed lightweight training framework, STLV-Align (Synergistic Touch-Language-Vision Alignment), achieving effective semantic alignment with minimal parameter adjustments (1%). Project Page: https://xiaoen0.github.io/touch.page/.

Tao Yang, You Li, Cheng Zhao et al.

In the past ten years, the use of 3D Time-of-Flight (ToF) LiDARs in mobile robotics has grown rapidly. Based on our accumulation of relevant research, this article systematically reviews and analyzes the use 3D ToF LiDARs in research and industrial applications. The former includes object detection, robot localization, long-term autonomy, LiDAR data processing under adverse weather conditions, and sensor fusion. The latter encompasses service robots, assisted and autonomous driving, and recent applications performed in response to public health crises. We hope that our efforts can effectively provide readers with relevant references and promote the deployment of existing mature technologies in real-world systems.

Hailiang Tang, Xiaoji Niu, Tisheng Zhang et al.

Odometer has been proven to significantly improve the accuracy of the Global Navigation Satellite System / Inertial Navigation System (GNSS/INS) integrated vehicle navigation in GNSS-challenged environments. However, the odometer is inaccessible in many applications, especially for aftermarket devices. To apply forward speed aiding without hardware wheeled odometer, we propose OdoNet, an untethered one-dimensional Convolution Neural Network (CNN)-based pseudo-odometer model learning from a single Inertial Measurement Unit (IMU), which can act as an alternative to the wheeled odometer. Dedicated experiments have been conducted to verify the feasibility and robustness of the OdoNet. The results indicate that the IMU individuality, the vehicle loads, and the road conditions have little impact on the robustness and precision of the OdoNet, while the IMU biases and the mounting angles may notably ruin the OdoNet. Thus, a data-cleaning procedure is added to effectively mitigate the impacts of the IMU biases and the mounting angles. Compared to the process using only non-holonomic constraint (NHC), after employing the pseudo-odometer, the positioning error is reduced by around 68%, while the percentage is around 74% for the hardware wheeled odometer. In conclusion, the proposed OdoNet can be employed as an untethered pseudo-odometer for vehicle navigation, which can efficiently improve the accuracy and reliability of the positioning in GNSS-denied environments.

Edouard Capellier, Franck Davoine, Veronique Cherfaoui et al.

LIDAR sensors are usually used to provide autonomous vehicles with 3D representations of their environment. In ideal conditions, geometrical models could detect the road in LIDAR scans, at the cost of a manual tuning of numerical constraints, and a lack of flexibility. We instead propose an evidential pipeline, to accumulate road detection results obtained from neural networks. First, we introduce RoadSeg, a new convolutional architecture that is optimized for road detection in LIDAR scans. RoadSeg is used to classify individual LIDAR points as either belonging to the road, or not. Yet, such point-level classification results need to be converted into a dense representation, that can be used by an autonomous vehicle. We thus secondly present an evidential road mapping algorithm, that fuses consecutive road detection results. We benefitted from a reinterpretation of logistic classifiers, which can be seen as generating a collection of simple evidential mass functions. An evidential grid map that depicts the road can then be obtained, by projecting the classification results from RoadSeg into grid cells, and by handling moving objects via conflict analysis. The system was trained and evaluated on real-life data. A python implementation maintains a 10 Hz framerate. Since road labels were needed for training, a soft labelling procedure, relying lane-level HD maps, was used to generate coarse training and validation sets. An additional test set was manually labelled for evaluation purposes. So as to reach satisfactory results, the system fuses road detection results obtained from three variants of RoadSeg, processing different LIDAR features.

You Li, Binli Luo, Ning Gui

Low-dimension graph embeddings have proved extremely useful in various downstream tasks in large graphs, e.g., link-related content recommendation and node classification tasks, etc. Most existing embedding approaches take nodes as the basic unit for information aggregation, e.g., node perception fields in GNN or con-textual nodes in random walks. The main drawback raised by such node-view is its lack of support for expressing the compound relationships between nodes, which results in the loss of a certain degree of graph information during embedding. To this end, this paper pro-poses PairE(Pair Embedding), a solution to use "pair", a higher level unit than a "node" as the core for graph embeddings. Accordingly, a multi-self-supervised auto-encoder is designed to fulfill two pretext tasks, to reconstruct the feature distribution for respective pairs and their surrounding context. PairE has three major advantages: 1) Informative, embedding beyond node-view are capable to preserve richer information of the graph; 2) Simple, the solutions provided by PairE are time-saving, storage-efficient, and require the fewer hyper-parameters; 3) High adaptability, with the introduced translator operator to map pair embeddings to the node embeddings, PairE can be effectively used in both the link-based and the node-based graph analysis. Experiment results show that PairE consistently outperforms the state of baselines in all four downstream tasks, especially with significant edges in the link-prediction and multi-label node classification tasks.

Shuyu Kong, You Li, Jia Wang et al.

Supervised learning on Deep Neural Networks (DNNs) is data hungry. Optimizing performance of DNN in the presence of noisy labels has become of paramount importance since collecting a large dataset will usually bring in noisy labels. Inspired by the robustness of K-Nearest Neighbors (KNN) against data noise, in this work, we propose to apply deep KNN for label cleanup. Our approach leverages DNNs for feature extraction and KNN for ground-truth label inference. We iteratively train the neural network and update labels to simultaneously proceed towards higher label recovery rate and better classification performance. Experiment results show that under the same setting, our approach outperforms existing label correction methods and achieves better accuracy on multiple datasets, e.g.,76.78% on Clothing1M dataset.

You Li, Zhuowen Lin

Attempts to use generative models for music generation have been common in recent years, and some of them have achieved good results. Pieces generated by some of these models are almost indistinguishable from those being composed by human composers. However, the research on the evaluation system for machine-generated music is still at a relatively early stage, and there is no uniform standard for such tasks. This paper proposes a stacked-LSTM binary classifier based on a language model, which can be used to distinguish the human composer's work from the machine-generated melody by learning the MIDI file's pitch, position, and duration.

You Li, Clément Le Bihan, Txomin Pourtau et al.

LiDARs are usually more accurate than cameras in distance measuring. Hence, there is strong interest to apply LiDARs in autonomous driving. Different existing approaches process the rich 3D point clouds for object detection, tracking and recognition. These methods generally require two initial steps: (1) filter points on the ground plane and (2) cluster non-ground points into objects. This paper proposes a field-tested fast 3D point cloud segmentation method for these two steps. Our specially designed algorithms allow instantly process raw LiDAR data packets, which significantly reduce the processing delay. In our tests on Velodyne UltraPuck, a 32 layers spinning LiDAR, the processing delay of clustering all the $360^\circ$ LiDAR measures is less than 1ms. Meanwhile, a coarse-to-fine scheme is applied to ensure the clustering quality. Our field experiments in public roads have shown that the proposed method significantly improves the speed of 3D point cloud clustering whilst maintains good accuracy.

Abdelhak Loukkal, Yves Grandvalet, Tom Drummond et al.

Camera-based end-to-end driving neural networks bring the promise of a low-cost system that maps camera images to driving control commands. These networks are appealing because they replace laborious hand engineered building blocks but their black-box nature makes them difficult to delve in case of failure. Recent works have shown the importance of using an explicit intermediate representation that has the benefits of increasing both the interpretability and the accuracy of networks' decisions. Nonetheless, these camera-based networks reason in camera view where scale is not homogeneous and hence not directly suitable for motion forecasting. In this paper, we introduce a novel monocular camera-only holistic end-to-end trajectory planning network with a Bird-Eye-View (BEV) intermediate representation that comes in the form of binary Occupancy Grid Maps (OGMs). To ease the prediction of OGMs in BEV from camera images, we introduce a novel scheme where the OGMs are first predicted as semantic masks in camera view and then warped in BEV using the homography between the two planes. The key element allowing this transformation to be applied to 3D objects such as vehicles, consists in predicting solely their footprint in camera-view, hence respecting the flat world hypothesis implied by the homography.

You Li, Ruizhi Chen, Xiaoji Niu et al.

The inertial navigation system (INS) has been widely used to provide self-contained and continuous motion estimation in intelligent transportation systems. Recently, the emergence of chip-level inertial sensors has expanded the relevant applications from positioning, navigation, and mobile mapping to location-based services, unmanned systems, and transportation big data. Meanwhile, benefit from the emergence of big data and the improvement of algorithms and computing power, artificial intelligence (AI) has become a consensus tool that has been successfully applied in various fields. This article reviews the research on using AI technology to enhance inertial sensing from various aspects, including sensor design and selection, calibration and error modeling, navigation and motion-sensing algorithms, multi-sensor information fusion, system evaluation, and practical application. Based on the over 30 representative articles selected from the nearly 300 related publications, this article summarizes the state of the art, advantages, and challenges on each aspect. Finally, it summarizes nine advantages and nine challenges of AI-enhanced inertial sensing and then points out future research directions.

You Li, Javier Ibanez-Guzman

Autonomous vehicles rely on their perception systems to acquire information about their immediate surroundings. It is necessary to detect the presence of other vehicles, pedestrians and other relevant entities. Safety concerns and the need for accurate estimations have led to the introduction of Light Detection and Ranging (LiDAR) systems in complement to the camera or radar-based perception systems. This article presents a review of state-of-the-art automotive LiDAR technologies and the perception algorithms used with those technologies. LiDAR systems are introduced first by analyzing the main components, from laser transmitter to its beam scanning mechanism. Advantages/disadvantages and the current status of various solutions are introduced and compared. Then, the specific perception pipeline for LiDAR data processing, from an autonomous vehicle perspective is detailed. The model-driven approaches and the emerging deep learning solutions are reviewed. Finally, we provide an overview of the limitations, challenges and trends for automotive LiDARs and perception systems.

You Li, Xin Hu, Yuan Zhuang et al.

Location is key to spatialize internet-of-things (IoT) data. However, it is challenging to use low-cost IoT devices for robust unsupervised localization (i.e., localization without training data that have known location labels). Thus, this paper proposes a deep reinforcement learning (DRL) based unsupervised wireless-localization method. The main contributions are as follows. (1) This paper proposes an approach to model a continuous wireless-localization process as a Markov decision process (MDP) and process it within a DRL framework. (2) To alleviate the challenge of obtaining rewards when using unlabeled data (e.g., daily-life crowdsourced data), this paper presents a reward-setting mechanism, which extracts robust landmark data from unlabeled wireless received signal strengths (RSS). (3) To ease requirements for model re-training when using DRL for localization, this paper uses RSS measurements together with agent location to construct DRL inputs. The proposed method was tested by using field testing data from multiple Bluetooth 5 smart ear tags in a pasture. Meanwhile, the experimental verification process reflected the advantages and challenges for using DRL in wireless localization.

You Li, Pierre Duthon, Michèle Colomb et al.

This article focuses on analyzing the performance of a typical time-of-flight (ToF) LiDAR under fog environment. By controlling the fog density within CEREMA Adverse Weather Facility 1 , the relations between the ranging performance and fogs are both qualitatively and quantitatively investigated. Furthermore, based on the collected data, a machine learning based model is trained to predict the minimum fog visibility that allows successful ranging for this type of LiDAR. The revealed experimental results and methods are helpful for ToF LiDAR specifications from automotive industry.