Le Li

Meituan LongCat Team, Xunliang Cai, Meng Cheng et al.

Despite advances in audio-driven video generation, achieving commercial-grade stability remains challenging. We present LongCat-Video-Avatar 1.5, an upgraded open-source framework prioritizing systematic engineering and production-readiness over architectural novelty. By upgrading the audio encoder to Whisper Large and meticulously scaling our training recipes, v1.5 achieves accurate lip-synchronization, full-body temporal stability, and robust long-video generation with strict identity consistency. Through rigorous data curation and RLHF Training, the model readily generalizes to stylized domains such as anime and animals, and natively handles complex real-world conditions, such as multi-person interactions and object handling. Furthermore, addressing the practical demands of industrial deployment, we employ advanced step distillation to accelerate inference to an optimal 8 NFE, achieving a favorable trade-off between serving efficiency and visual fidelity. The superiority of our approach is validated through extensive quantitative metrics and a rigorous human evaluation conducted on a comprehensive benchmark of over 500 diverse test cases. Results show that v1.5 achieves competitive or superior performance compared to leading closed-source systems (e.g., HeyGen, OmniHuman 1.5, Kling Avatar 2.0) across human-likeness ratings and expert-level quality assessments on our benchmark. With its open-source release, LongCat-Video-Avatar 1.5 narrows the gap between academic research prototypes and commercial-grade deployment.

Le Li, Jiale Wei, Pai Peng et al.

Data scarcity and data imbalance have attracted a lot of attention in many fields. Data augmentation, explored as an effective approach to tackle them, can improve the robustness and efficiency of classification models by generating new samples. This paper presents REPRINT, a simple and effective hidden-space data augmentation method for imbalanced data classification. Given hidden-space representations of samples in each class, REPRINT extrapolates, in a randomized fashion, augmented examples for target class by using subspaces spanned by principal components to summarize distribution structure of both source and target class. Consequently, the examples generated would diversify the target while maintaining the original geometry of target distribution. Besides, this method involves a label refinement component which allows to synthesize new soft labels for augmented examples. Compared with different NLP data augmentation approaches under a range of data imbalanced scenarios on four text classification benchmark, REPRINT shows prominent improvements. Moreover, through comprehensive ablation studies, we show that label refinement is better than label-preserving for augmented examples, and that our method suggests stable and consistent improvements in terms of suitable choices of principal components. Moreover, REPRINT is appealing for its easy-to-use since it contains only one hyperparameter determining the dimension of subspace and requires low computational resource.

Morunliu Yang, Ruotao Xu, Le Li et al.

Omnimodal large language models (OmniLLMs) have recently gained increasing attention for unified audio-video understanding. However, processing long multimodal token sequences introduces substantial computational overhead, making efficient token compression crucial. Existing methods typically rely on fixed, modality-specific guidance, which fails to account for the varying importance of modalities across different queries. To address this limitation, we propose $\textbf{OmniSelect}$, a training-free, modality-adaptive token pruning framework that dynamically selects appropriate compression strategies for multimodal inputs. Specifically, we leverage a lightweight AudioCLIP model to estimate cross-modal relevance and categorize each input into three pruning regimes: Audio-Centric, Video-Centric, and Uniform pruning. Based on these relevance scores, OmniSelect further performs fine-grained token pruning within each temporal group, adaptively allocating pruning ratios to preserve informative tokens across modalities. By explicitly modeling modality preference and enabling dynamic strategy selection, OmniSelect effectively avoids the pitfalls of one-size-fits-all compression. Extensive experiments demonstrate that our method achieves efficient multimodal token reduction while maintaining strong performance, without requiring any additional training.

Jiawei Wang, Kai Wang, Shaojie Lin et al.

With the rapid advancement of Large Language Models (LLMs), LLM-based autonomous agents have shown the potential to function as digital employees, such as digital analysts, teachers, and programmers. In this paper, we develop an application-level testbed based on the open-source strategy game "Unciv", which has millions of active players, to enable researchers to build a "data flywheel" for studying human-like agents in the "digital players" task. This "Civilization"-like game features expansive decision-making spaces along with rich linguistic interactions such as diplomatic negotiations and acts of deception, posing significant challenges for LLM-based agents in terms of numerical reasoning and long-term planning. Another challenge for "digital players" is to generate human-like responses for social interaction, collaboration, and negotiation with human players. The open-source project can be found at https:/github.com/fuxiAIlab/CivAgent.

Zhe Kong, Le Li, Yong Zhang et al.

Real-world video super-resolution (VSR) presents significant challenges due to complex and unpredictable degradations. Although some recent methods utilize image diffusion models for VSR and have shown improved detail generation capabilities, they still struggle to produce temporally consistent frames. We attempt to use Stable Video Diffusion (SVD) combined with ControlNet to address this issue. However, due to the intrinsic image-animation characteristics of SVD, it is challenging to generate fine details using only low-quality videos. To tackle this problem, we propose DAM-VSR, an appearance and motion disentanglement framework for VSR. This framework disentangles VSR into appearance enhancement and motion control problems. Specifically, appearance enhancement is achieved through reference image super-resolution, while motion control is achieved through video ControlNet. This disentanglement fully leverages the generative prior of video diffusion models and the detail generation capabilities of image super-resolution models. Furthermore, equipped with the proposed motion-aligned bidirectional sampling strategy, DAM-VSR can conduct VSR on longer input videos. DAM-VSR achieves state-of-the-art performance on real-world data and AIGC data, demonstrating its powerful detail generation capabilities.

Bihan Xu, Shiwei Zhao, Runze Wu et al.

Within the domain of Massively Multiplayer Online (MMO) economy research, Agent-Based Modeling (ABM) has emerged as a robust tool for analyzing game economics, evolving from rule-based agents to decision-making agents enhanced by reinforcement learning. Nevertheless, existing works encounter significant challenges when attempting to emulate human-like economic activities among agents, particularly regarding agent reliability, sociability, and interpretability. In this study, we take a preliminary step in introducing a novel approach using Large Language Models (LLMs) in MMO economy simulation. Leveraging LLMs' role-playing proficiency, generative capacity, and reasoning aptitude, we design LLM-driven agents with human-like decision-making and adaptability. These agents are equipped with the abilities of role-playing, perception, memory, and reasoning, addressing the aforementioned challenges effectively. Simulation experiments focusing on in-game economic activities demonstrate that LLM-empowered agents can promote emergent phenomena like role specialization and price fluctuations in line with market rules.

Yinting Wu, Pai Peng, Bo Cai et al.

Adversarial training methods commonly generate independent initial perturbation for adversarial samples from a simple uniform distribution, and obtain the training batch for the classifier without selection. In this work, we propose a simple yet effective training framework called Batch-in-Batch (BB) to enhance models robustness. It involves specifically a joint construction of initial values that could simultaneously generates $m$ sets of perturbations from the original batch set to provide more diversity for adversarial samples; and also includes various sample selection strategies that enable the trained models to have smoother losses and avoid overconfident outputs. Through extensive experiments on three benchmark datasets (CIFAR-10, SVHN, CIFAR-100) with two networks (PreActResNet18 and WideResNet28-10) that are used in both the single-step (Noise-Fast Gradient Sign Method, N-FGSM) and multi-step (Projected Gradient Descent, PGD-10) adversarial training, we show that models trained within the BB framework consistently have higher adversarial accuracy across various adversarial settings, notably achieving over a 13% improvement on the SVHN dataset with an attack radius of 8/255 compared to the N-FGSM baseline model. Furthermore, experimental analysis of the efficiency of both the proposed initial perturbation method and sample selection strategies validates our insights. Finally, we show that our framework is cost-effective in terms of computational resources, even with a relatively large value of $m$.

Rongsheng Zhang, Xiaoxi Mao, Le Li et al.

Recently, a variety of neural models have been proposed for lyrics generation. However, most previous work completes the generation process in a single pass with little human intervention. We believe that lyrics creation is a creative process with human intelligence centered. AI should play a role as an assistant in the lyrics creation process, where human interactions are crucial for high-quality creation. This paper demonstrates \textit{Youling}, an AI-assisted lyrics creation system, designed to collaborate with music creators. In the lyrics generation process, \textit{Youling} supports traditional one pass full-text generation mode as well as an interactive generation mode, which allows users to select the satisfactory sentences from generated candidates conditioned on preceding context. The system also provides a revision module which enables users to revise undesired sentences or words of lyrics repeatedly. Besides, \textit{Youling} allows users to use multifaceted attributes to control the content and format of generated lyrics. The demo video of the system is available at https://youtu.be/DFeNpHk0pm4.

Zhilei Liu, Yunpeng Wu, Le Li et al.

Previous research on face restoration often focused on repairing a specific type of low-quality facial images such as low-resolution (LR) or occluded facial images. However, in the real world, both the above-mentioned forms of image degradation often coexist. Therefore, it is important to design a model that can repair LR occluded images simultaneously. This paper proposes a multi-scale feature graph generative adversarial network (MFG-GAN) to implement the face restoration of images in which both degradation modes coexist, and also to repair images with a single type of degradation. Based on the GAN, the MFG-GAN integrates the graph convolution and feature pyramid network to restore occluded low-resolution face images to non-occluded high-resolution face images. The MFG-GAN uses a set of customized losses to ensure that high-quality images are generated. In addition, we designed the network in an end-to-end format. Experimental results on the public-domain CelebA and Helen databases show that the proposed approach outperforms state-of-the-art methods in performing face super-resolution (up to 4x or 8x) and face completion simultaneously. Cross-database testing also revealed that the proposed approach has good generalizability.

Yong Zhang, Le Li, Zhilei Liu et al.

Kinship face synthesis is an interesting topic raised to answer questions like "what will your future children look like?". Published approaches to this topic are limited. Most of the existing methods train models for one-versus-one kin relation, which only consider one parent face and one child face by directly using an auto-encoder without any explicit control over the resemblance of the synthesized face to the parent face. In this paper, we propose a novel method for controllable descendant face synthesis, which models two-versus-one kin relation between two parent faces and one child face. Our model consists of an inheritance module and an attribute enhancement module, where the former is designed for accurate control over the resemblance between the synthesized face and parent faces, and the latter is designed for control over age and gender. As there is no large scale database with father-mother-child kinship annotation, we propose an effective strategy to train the model without using the ground truth descendant faces. No carefully designed image pairs are required for learning except only age and gender labels of training faces. We conduct comprehensive experimental evaluations on three public benchmark databases, which demonstrates encouraging results.

Zhilei Liu, Le Li, Yunpeng Wu et al.

Facial expression analysis in the wild is challenging when the facial image is with low resolution or partial occlusion. Considering the correlations among different facial local regions under different facial expressions, this paper proposes a novel facial expression restoration method based on generative adversarial network by integrating an improved graph convolutional network (IGCN) and region relation modeling block (RRMB). Unlike conventional graph convolutional networks taking vectors as input features, IGCN can use tensors of face patches as inputs. It is better to retain the structure information of face patches. The proposed RRMB is designed to address facial generative tasks including inpainting and super-resolution with facial action units detection, which aims to restore facial expression as the ground-truth. Extensive experiments conducted on BP4D and DISFA benchmarks demonstrate the effectiveness of our proposed method through quantitative and qualitative evaluations.

Benjamin Guedj, Le Li

When confronted with massive data streams, summarizing data with dimension reduction methods such as PCA raises theoretical and algorithmic pitfalls. Principal curves act as a nonlinear generalization of PCA and the present paper proposes a novel algorithm to automatically and sequentially learn principal curves from data streams. We show that our procedure is supported by regret bounds with optimal sublinear remainder terms. A greedy local search implementation (called \texttt{slpc}, for Sequential Learning Principal Curves) that incorporates both sleeping experts and multi-armed bandit ingredients is presented, along with its regret computation and performance on synthetic and real-life data.

Le Li, Benjamin Guedj, Sébastien Loustau

When faced with high frequency streams of data, clustering raises theoretical and algorithmic pitfalls. We introduce a new and adaptive online clustering algorithm relying on a quasi-Bayesian approach, with a dynamic (i.e., time-dependent) estimation of the (unknown and changing) number of clusters. We prove that our approach is supported by minimax regret bounds. We also provide an RJMCMC-flavored implementation (called PACBO, see https://cran.r-project.org/web/packages/PACBO/index.html) for which we give a convergence guarantee. Finally, numerical experiments illustrate the potential of our procedure.

Le Li, Jianjun Yang, Yang Xu et al.

Nonnegative matrix factorization (NMF) has been successfully applied to many areas for classification and clustering. Commonly-used NMF algorithms mainly target on minimizing the $l_2$ distance or Kullback-Leibler (KL) divergence, which may not be suitable for nonlinear case. In this paper, we propose a new decomposition method by maximizing the correntropy between the original and the product of two low-rank matrices for document clustering. This method also allows us to learn the new basis vectors of the semantic feature space from the data. To our knowledge, we haven't seen any work has been done by maximizing correntropy in NMF to cluster high dimensional document data. Our experiment results show the supremacy of our proposed method over other variants of NMF algorithm on Reuters21578 and TDT2 databasets.

Le Li, Jianjun Yang, Kaili Zhao et al.

Non-negative matrix factorization (NMF) has proved effective in many clustering and classification tasks. The classic ways to measure the errors between the original and the reconstructed matrix are $l_2$ distance or Kullback-Leibler (KL) divergence. However, nonlinear cases are not properly handled when we use these error measures. As a consequence, alternative measures based on nonlinear kernels, such as correntropy, are proposed. However, the current correntropy-based NMF only targets on the low-level features without considering the intrinsic geometrical distribution of data. In this paper, we propose a new NMF algorithm that preserves local invariance by adding graph regularization into the process of max-correntropy-based matrix factorization. Meanwhile, each feature can learn corresponding kernel from the data. The experiment results of Caltech101 and Caltech256 show the benefits of such combination against other NMF algorithms for the unsupervised image clustering.

Yuzhu Zhou, Le Li, Honggang Zhang

In this paper, we present a region-based pLSA model to accomplish the task of total scene annotation. To be more specific, we not only properly generate a list of tags for each image, but also localizing each region with its corresponding tag. We integrate advantages of different existing region-based works: employ efficient and powerful JSEG algorithm for segmentation so that each region can easily express meaningful object information; the introduction of pLSA model can help better capturing semantic information behind the low-level features. Moreover, we also propose an adaptive padding mechanism to automatically choose the optimal padding strategy for each region, which directly increases the overall system performance. Finally we conduct 3 experiments to verify our ideas on Corel database and demonstrate the effectiveness and accuracy of our system.