Yong-Jin Liu

Andrew Zhao, Matthieu Gaetan Lin, Yangguang Li et al. · tsinghua

Unsupervised reinforcement learning aims at learning a generalist policy in a reward-free manner for fast adaptation to downstream tasks. Most of the existing methods propose to provide an intrinsic reward based on surprise. Maximizing or minimizing surprise drives the agent to either explore or gain control over its environment. However, both strategies rely on a strong assumption: the entropy of the environment's dynamics is either high or low. This assumption may not always hold in real-world scenarios, where the entropy of the environment's dynamics may be unknown. Hence, choosing between the two objectives is a dilemma. We propose a novel yet simple mixture of policies to address this concern, allowing us to optimize an objective that simultaneously maximizes and minimizes the surprise. Concretely, we train one mixture component whose objective is to maximize the surprise and another whose objective is to minimize the surprise. Hence, our method does not make assumptions about the entropy of the environment's dynamics. We call our method a $\textbf{M}\text{ixture }\textbf{O}\text{f }\textbf{S}\text{urprise}\textbf{S}$ (MOSS) for unsupervised reinforcement learning. Experimental results show that our simple method achieves state-of-the-art performance on the URLB benchmark, outperforming previous pure surprise maximization-based objectives. Our code is available at: https://github.com/LeapLabTHU/MOSS.

Wang Zhao, Shaohui Liu, Hengkai Guo et al.

Estimating the pose of a moving camera from monocular video is a challenging problem, especially due to the presence of moving objects in dynamic environments, where the performance of existing camera pose estimation methods are susceptible to pixels that are not geometrically consistent. To tackle this challenge, we present a robust dense indirect structure-from-motion method for videos that is based on dense correspondence initialized from pairwise optical flow. Our key idea is to optimize long-range video correspondence as dense point trajectories and use it to learn robust estimation of motion segmentation. A novel neural network architecture is proposed for processing irregular point trajectory data. Camera poses are then estimated and optimized with global bundle adjustment over the portion of long-range point trajectories that are classified as static. Experiments on MPI Sintel dataset show that our system produces significantly more accurate camera trajectories compared to existing state-of-the-art methods. In addition, our method is able to retain reasonable accuracy of camera poses on fully static scenes, which consistently outperforms strong state-of-the-art dense correspondence based methods with end-to-end deep learning, demonstrating the potential of dense indirect methods based on optical flow and point trajectories. As the point trajectory representation is general, we further present results and comparisons on in-the-wild monocular videos with complex motion of dynamic objects. Code is available at https://github.com/bytedance/particle-sfm.

Andrew Zhao, Daniel Huang, Quentin Xu et al. · tsinghua

The recent surge in research interest in applying large language models (LLMs) to decision-making tasks has flourished by leveraging the extensive world knowledge embedded in LLMs. While there is a growing demand to tailor LLMs for custom decision-making tasks, finetuning them for specific tasks is resource-intensive and may diminish the model's generalization capabilities. Moreover, state-of-the-art language models like GPT-4 and Claude are primarily accessible through API calls, with their parametric weights remaining proprietary and unavailable to the public. This scenario emphasizes the growing need for new methodologies that allow learning from agent experiences without requiring parametric updates. To address these problems, we introduce the Experiential Learning (ExpeL) agent. Our agent autonomously gathers experiences and extracts knowledge using natural language from a collection of training tasks. At inference, the agent recalls its extracted insights and past experiences to make informed decisions. Our empirical results highlight the robust learning efficacy of the ExpeL agent, indicating a consistent enhancement in its performance as it accumulates experiences. We further explore the emerging capabilities and transfer learning potential of the ExpeL agent through qualitative observations and additional experiments.

Yubin Hu, Yuze He, Yanghao Li et al.

Video semantic segmentation (VSS) is a computationally expensive task due to the per-frame prediction for videos of high frame rates. In recent work, compact models or adaptive network strategies have been proposed for efficient VSS. However, they did not consider a crucial factor that affects the computational cost from the input side: the input resolution. In this paper, we propose an altering resolution framework called AR-Seg for compressed videos to achieve efficient VSS. AR-Seg aims to reduce the computational cost by using low resolution for non-keyframes. To prevent the performance degradation caused by downsampling, we design a Cross Resolution Feature Fusion (CReFF) module, and supervise it with a novel Feature Similarity Training (FST) strategy. Specifically, CReFF first makes use of motion vectors stored in a compressed video to warp features from high-resolution keyframes to low-resolution non-keyframes for better spatial alignment, and then selectively aggregates the warped features with local attention mechanism. Furthermore, the proposed FST supervises the aggregated features with high-resolution features through an explicit similarity loss and an implicit constraint from the shared decoding layer. Extensive experiments on CamVid and Cityscapes show that AR-Seg achieves state-of-the-art performance and is compatible with different segmentation backbones. On CamVid, AR-Seg saves 67% computational cost (measured in GFLOPs) with the PSPNet18 backbone while maintaining high segmentation accuracy. Code: https://github.com/THU-LYJ-Lab/AR-Seg.

Sheng Ye, Yubin Hu, Matthieu Lin et al.

The reconstruction of indoor scenes from multi-view RGB images is challenging due to the coexistence of flat and texture-less regions alongside delicate and fine-grained regions. Recent methods leverage neural radiance fields aided by predicted surface normal priors to recover the scene geometry. These methods excel in producing complete and smooth results for floor and wall areas. However, they struggle to capture complex surfaces with high-frequency structures due to the inadequate neural representation and the inaccurately predicted normal priors. This work aims to reconstruct high-fidelity surfaces with fine-grained details by addressing the above limitations. To improve the capacity of the implicit representation, we propose a hybrid architecture to represent low-frequency and high-frequency regions separately. To enhance the normal priors, we introduce a simple yet effective image sharpening and denoising technique, coupled with a network that estimates the pixel-wise uncertainty of the predicted surface normal vectors. Identifying such uncertainty can prevent our model from being misled by unreliable surface normal supervisions that hinder the accurate reconstruction of intricate geometries. Experiments on the benchmark datasets show that our method outperforms existing methods in terms of reconstruction quality. Furthermore, the proposed method also generalizes well to real-world indoor scenarios captured by our hand-held mobile phones. Our code is publicly available at: https://github.com/yec22/Fine-Grained-Indoor-Recon.

Nan Ma, Mohan Wang, Yiheng Han et al.

Cross-modality point cloud registration is confronted with significant challenges due to inherent differences in modalities between different sensors. We propose a cross-modality point cloud registration framework FF-LOGO: a cross-modality point cloud registration method with feature filtering and local-global optimization. The cross-modality feature correlation filtering module extracts geometric transformation-invariant features from cross-modality point clouds and achieves point selection by feature matching. We also introduce a cross-modality optimization process, including a local adaptive key region aggregation module and a global modality consistency fusion optimization module. Experimental results demonstrate that our two-stage optimization significantly improves the registration accuracy of the feature association and selection module. Our method achieves a substantial increase in recall rate compared to the current state-of-the-art methods on the 3DCSR dataset, improving from 40.59% to 75.74%. Our code will be available at https://github.com/wangmohan17/FFLOGO.

Yubin Hu, Sheng Ye, Wang Zhao et al.

Occlusion is a common issue in 3D reconstruction from RGB-D videos, often blocking the complete reconstruction of objects and presenting an ongoing problem. In this paper, we propose a novel framework, empowered by a 2D diffusion-based in-painting model, to reconstruct complete surfaces for the hidden parts of objects. Specifically, we utilize a pre-trained diffusion model to fill in the hidden areas of 2D images. Then we use these in-painted images to optimize a neural implicit surface representation for each instance for 3D reconstruction. Since creating the in-painting masks needed for this process is tricky, we adopt a human-in-the-loop strategy that involves very little human engagement to generate high-quality masks. Moreover, some parts of objects can be totally hidden because the videos are usually shot from limited perspectives. To ensure recovering these invisible areas, we develop a cascaded network architecture for predicting signed distance field, making use of different frequency bands of positional encoding and maintaining overall smoothness. Besides the commonly used rendering loss, Eikonal loss, and silhouette loss, we adopt a CLIP-based semantic consistency loss to guide the surface from unseen camera angles. Experiments on ScanNet scenes show that our proposed framework achieves state-of-the-art accuracy and completeness in object-level reconstruction from scene-level RGB-D videos. Code: https://github.com/THU-LYJ-Lab/O2-Recon.

Yiheng Han, Irvin Haozhe Zhan, Long Zeng et al.

Some robust point cloud registration approaches with controllable pose refinement magnitude, such as ICP and its variants, are commonly used to improve 6D pose estimation accuracy. However, the effectiveness of these methods gradually diminishes with the advancement of deep learning techniques and the enhancement of initial pose accuracy, primarily due to their lack of specific design for pose refinement. In this paper, we propose Point Cloud Completion and Keypoint Refinement with Fusion Data (PCKRF), a new pose refinement pipeline for 6D pose estimation. The pipeline consists of two steps. First, it completes the input point clouds via a novel pose-sensitive point completion network. The network uses both local and global features with pose information during point completion. Then, it registers the completed object point cloud with the corresponding target point cloud by our proposed Color supported Iterative KeyPoint (CIKP) method. The CIKP method introduces color information into registration and registers a point cloud around each keypoint to increase stability. The PCKRF pipeline can be integrated with existing popular 6D pose estimation methods, such as the full flow bidirectional fusion network, to further improve their pose estimation accuracy. Experiments demonstrate that our method exhibits superior stability compared to existing approaches when optimizing initial poses with relatively high precision. Notably, the results indicate that our method effectively complements most existing pose estimation techniques, leading to improved performance in most cases. Furthermore, our method achieves promising results even in challenging scenarios involving textureless and symmetrical objects. Our source code is available at https://github.com/zhanhz/KRF.

Zipeng Ye, Zhiyao Sun, Yu-Hui Wen et al. · tsinghua

Recently, talking-face video generation has received considerable attention. So far most methods generate results with neutral expressions or expressions that are implicitly determined by neural networks in an uncontrollable way. In this paper, we propose a method to generate talking-face videos with continuously controllable expressions in real-time. Our method is based on an important observation: In contrast to facial geometry of moderate resolution, most expression information lies in textures. Then we make use of neural textures to generate high-quality talking face videos and design a novel neural network that can generate neural textures for image frames (which we called dynamic neural textures) based on the input expression and continuous intensity expression coding (CIEC). Our method uses 3DMM as a 3D model to sample the dynamic neural texture. The 3DMM does not cover the teeth area, so we propose a teeth submodule to complete the details in teeth. Results and an ablation study show the effectiveness of our method in generating high-quality talking-face videos with continuously controllable expressions. We also set up four baseline methods by combining existing representative methods and compare them with our method. Experimental results including a user study show that our method has the best performance.

Yuze He, Yushi Bai, Matthieu Lin et al. · tsinghua

Recent methods in text-to-3D leverage powerful pretrained diffusion models to optimize NeRF. Notably, these methods are able to produce high-quality 3D scenes without training on 3D data. Due to the open-ended nature of the task, most studies evaluate their results with subjective case studies and user experiments, thereby presenting a challenge in quantitatively addressing the question: How has current progress in Text-to-3D gone so far? In this paper, we introduce T$^3$Bench, the first comprehensive text-to-3D benchmark containing diverse text prompts of three increasing complexity levels that are specially designed for 3D generation. To assess both the subjective quality and the text alignment, we propose two automatic metrics based on multi-view images produced by the 3D contents. The quality metric combines multi-view text-image scores and regional convolution to detect quality and view inconsistency. The alignment metric uses multi-view captioning and GPT-4 evaluation to measure text-3D consistency. Both metrics closely correlate with different dimensions of human judgments, providing a paradigm for efficiently evaluating text-to-3D models. The benchmarking results, shown in Fig. 1, reveal performance differences among an extensive 10 prevalent text-to-3D methods. Our analysis further highlights the common struggles for current methods on generating surroundings and multi-object scenes, as well as the bottleneck of leveraging 2D guidance for 3D generation. Our project page is available at: https://t3bench.com.

Aihua Mao, Zihui Du, Yu-Hui Wen et al.

Point cloud denoising aims to restore clean point clouds from raw observations corrupted by noise and outliers while preserving the fine-grained details. We present a novel deep learning-based denoising model, that incorporates normalizing flows and noise disentanglement techniques to achieve high denoising accuracy. Unlike existing works that extract features of point clouds for point-wise correction, we formulate the denoising process from the perspective of distribution learning and feature disentanglement. By considering noisy point clouds as a joint distribution of clean points and noise, the denoised results can be derived from disentangling the noise counterpart from latent point representation, and the mapping between Euclidean and latent spaces is modeled by normalizing flows. We evaluate our method on synthesized 3D models and real-world datasets with various noise settings. Qualitative and quantitative results show that our method outperforms previous state-of-the-art deep learning-based approaches.

Mengfei Xia, Yujun Shen, Ceyuan Yang et al.

Generative adversarial networks (GANs) usually struggle in learning from highly diverse data, whose underlying manifold is complex. In this work, we revisit the mathematical foundations of GANs, and theoretically reveal that the native adversarial loss for GAN training is insufficient to fix the problem of \textit{subsets with positive Lebesgue measure of the generated data manifold lying out of the real data manifold}. Instead, we find that score matching serves as a promising solution to this issue thanks to its capability of persistently pushing the generated data points towards the real data manifold. We thereby propose to improve the optimization of GANs with score matching regularity (SMaRt). Regarding the empirical evidences, we first design a toy example to show that training GANs by the aid of a ground-truth score function can help reproduce the real data distribution more accurately, and then confirm that our approach can consistently boost the synthesis performance of various state-of-the-art GANs on real-world datasets with pre-trained diffusion models acting as the approximate score function. For instance, when training Aurora on the ImageNet $64\times64$ dataset, we manage to improve FID from 8.87 to 7.11, on par with the performance of one-step consistency model. Code is available at \href{https://github.com/thuxmf/SMaRt}{https://github.com/thuxmf/SMaRt}.

Zhiyao Sun, Yu-Hui Wen, Tian Lv et al.

Recently audio-driven talking face video generation has attracted considerable attention. However, very few researches address the issue of emotional editing of these talking face videos with continuously controllable expressions, which is a strong demand in the industry. The challenge is that speech-related expressions and emotion-related expressions are often highly coupled. Meanwhile, traditional image-to-image translation methods cannot work well in our application due to the coupling of expressions with other attributes such as poses, i.e., translating the expression of the character in each frame may simultaneously change the head pose due to the bias of the training data distribution. In this paper, we propose a high-quality facial expression editing method for talking face videos, allowing the user to control the target emotion in the edited video continuously. We present a new perspective for this task as a special case of motion information editing, where we use a 3DMM to capture major facial movements and an associated texture map modeled by a StyleGAN to capture appearance details. Both representations (3DMM and texture map) contain emotional information and can be continuously modified by neural networks and easily smoothed by averaging in coefficient/latent spaces, making our method simple yet effective. We also introduce a mouth shape preservation loss to control the trade-off between lip synchronization and the degree of exaggeration of the edited expression. Extensive experiments and a user study show that our method achieves state-of-the-art performance across various evaluation criteria.

Mengfei Xia, Yujun Shen, Changsong Lei et al.

A diffusion model, which is formulated to produce an image using thousands of denoising steps, usually suffers from a slow inference speed. Existing acceleration algorithms simplify the sampling by skipping most steps yet exhibit considerable performance degradation. By viewing the generation of diffusion models as a discretized integral process, we argue that the quality drop is partly caused by applying an inaccurate integral direction to a timestep interval. To rectify this issue, we propose a \textbf{timestep tuner} that helps find a more accurate integral direction for a particular interval at the minimum cost. Specifically, at each denoising step, we replace the original parameterization by conditioning the network on a new timestep, enforcing the sampling distribution towards the real one. Extensive experiments show that our plug-in design can be trained efficiently and boost the inference performance of various state-of-the-art acceleration methods, especially when there are few denoising steps. For example, when using 10 denoising steps on LSUN Bedroom dataset, we improve the FID of DDIM from 9.65 to 6.07, simply by adopting our method for a more appropriate set of timesteps. Code is available at \href{https://github.com/THU-LYJ-Lab/time-tuner}{https://github.com/THU-LYJ-Lab/time-tuner}.

Zhiyao Sun, Tian Lv, Sheng Ye et al.

The generation of stylistic 3D facial animations driven by speech presents a significant challenge as it requires learning a many-to-many mapping between speech, style, and the corresponding natural facial motion. However, existing methods either employ a deterministic model for speech-to-motion mapping or encode the style using a one-hot encoding scheme. Notably, the one-hot encoding approach fails to capture the complexity of the style and thus limits generalization ability. In this paper, we propose DiffPoseTalk, a generative framework based on the diffusion model combined with a style encoder that extracts style embeddings from short reference videos. During inference, we employ classifier-free guidance to guide the generation process based on the speech and style. In particular, our style includes the generation of head poses, thereby enhancing user perception. Additionally, we address the shortage of scanned 3D talking face data by training our model on reconstructed 3DMM parameters from a high-quality, in-the-wild audio-visual dataset. Extensive experiments and user study demonstrate that our approach outperforms state-of-the-art methods. The code and dataset are at https://diffposetalk.github.io .

Andrew Zhao, Erle Zhu, Rui Lu et al. · tsinghua

Humans possess the ability to draw on past experiences explicitly when learning new tasks and applying them accordingly. We believe this capacity for self-referencing is especially advantageous for reinforcement learning agents in the unsupervised pretrain-then-finetune setting. During pretraining, an agent's past experiences can be explicitly utilized to mitigate the nonstationarity of intrinsic rewards. In the finetuning phase, referencing historical trajectories prevents the unlearning of valuable exploratory behaviors. Motivated by these benefits, we propose the Self-Reference (SR) approach, an add-on module explicitly designed to leverage historical information and enhance agent performance within the pretrain-finetune paradigm. Our approach achieves state-of-the-art results in terms of Interquartile Mean (IQM) performance and Optimality Gap reduction on the Unsupervised Reinforcement Learning Benchmark for model-free methods, recording an 86% IQM and a 16% Optimality Gap. Additionally, it improves current algorithms by up to 17% IQM and reduces the Optimality Gap by 31%. Beyond performance enhancement, the Self-Reference add-on also increases sample efficiency, a crucial attribute for real-world applications.

Sheng Ye, Zhen-Hui Dong, Yubin Hu et al.

3D Gaussian Splatting has recently emerged as a powerful representation that can synthesize remarkable novel views using consistent multi-view images as input. However, we notice that images captured in dark environments where the scenes are not fully illuminated can exhibit considerable brightness variations and multi-view inconsistency, which poses great challenges to 3D Gaussian Splatting and severely degrades its performance. To tackle this problem, we propose Gaussian-DK. Observing that inconsistencies are mainly caused by camera imaging, we represent a consistent radiance field of the physical world using a set of anisotropic 3D Gaussians, and design a camera response module to compensate for multi-view inconsistencies. We also introduce a step-based gradient scaling strategy to constrain Gaussians near the camera, which turn out to be floaters, from splitting and cloning. Experiments on our proposed benchmark dataset demonstrate that Gaussian-DK produces high-quality renderings without ghosting and floater artifacts and significantly outperforms existing methods. Furthermore, we can also synthesize light-up images by controlling exposure levels that clearly show details in shadow areas.

Sheng Ye, Yuze He, Matthieu Lin et al.

Neural implicit representations have revolutionized dense multi-view surface reconstruction, yet their performance significantly diminishes with sparse input views. A few pioneering works have sought to tackle the challenge of sparse-view reconstruction by leveraging additional geometric priors or multi-scene generalizability. However, they are still hindered by the imperfect choice of input views, using images under empirically determined viewpoints to provide considerable overlap. We propose PVP-Recon, a novel and effective sparse-view surface reconstruction method that progressively plans the next best views to form an optimal set of sparse viewpoints for image capturing. PVP-Recon starts initial surface reconstruction with as few as 3 views and progressively adds new views which are determined based on a novel warping score that reflects the information gain of each newly added view. This progressive view planning progress is interleaved with a neural SDF-based reconstruction module that utilizes multi-resolution hash features, enhanced by a progressive training scheme and a directional Hessian loss. Quantitative and qualitative experiments on three benchmark datasets show that our framework achieves high-quality reconstruction with a constrained input budget and outperforms existing baselines.

Yuze He, Peng Wang, Yubin Hu et al.

This paper presents a flexible representation of neural radiance fields based on multi-plane images (MPI), for high-quality view synthesis of complex scenes. MPI with Normalized Device Coordinate (NDC) parameterization is widely used in NeRF learning for its simple definition, easy calculation, and powerful ability to represent unbounded scenes. However, existing NeRF works that adopt MPI representation for novel view synthesis can only handle simple forward-facing unbounded scenes, where the input cameras are all observing in similar directions with small relative translations. Hence, extending these MPI-based methods to more complex scenes like large-range or even 360-degree scenes is very challenging. In this paper, we explore the potential of MPI and show that MPI can synthesize high-quality novel views of complex scenes with diverse camera distributions and view directions, which are not only limited to simple forward-facing scenes. Our key idea is to encode the neural radiance field with multiple MPIs facing different directions and blend them with an adaptive blending operation. For each region of the scene, the blending operation gives larger blending weights to those advantaged MPIs with stronger local representation abilities while giving lower weights to those with weaker representation abilities. Such blending operation automatically modulates the multiple MPIs to appropriately represent the diverse local density and color information. Experiments on the KITTI dataset and ScanNet dataset demonstrate that our proposed MMPI synthesizes high-quality images from diverse camera pose distributions and is fast to train, outperforming the previous fast-training NeRF methods for novel view synthesis. Moreover, we show that MMPI can encode extremely long trajectories and produce novel view renderings, demonstrating its potential in applications like autonomous driving.

Mengfei Xia, Nan Xue, Yujun Shen et al.

Classifier-Free Guidance (CFG), which combines the conditional and unconditional score functions with two coefficients summing to one, serves as a practical technique for diffusion model sampling. Theoretically, however, denoising with CFG \textit{cannot} be expressed as a reciprocal diffusion process, which may consequently leave some hidden risks during use. In this work, we revisit the theory behind CFG and rigorously confirm that the improper configuration of the combination coefficients (\textit{i.e.}, the widely used summing-to-one version) brings about expectation shift of the generative distribution. To rectify this issue, we propose ReCFG with a relaxation on the guidance coefficients such that denoising with \method strictly aligns with the diffusion theory. We further show that our approach enjoys a \textbf{\textit{closed-form}} solution given the guidance strength. That way, the rectified coefficients can be readily pre-computed via traversing the observed data, leaving the sampling speed barely affected. Empirical evidence on real-world data demonstrate the compatibility of our post-hoc design with existing state-of-the-art diffusion models, including both class-conditioned ones (\textit{e.g.}, EDM2 on ImageNet) and text-conditioned ones (\textit{e.g.}, SD3 on CC12M), without any retraining. Code is available at \href{https://github.com/thuxmf/recfg}{https://github.com/thuxmf/recfg}.

Jenny Sheng, Matthieu Lin, Andrew Zhao et al. · tsinghua

This paper presents an exploration of preference learning in text-to-motion generation. We find that current improvements in text-to-motion generation still rely on datasets requiring expert labelers with motion capture systems. Instead, learning from human preference data does not require motion capture systems; a labeler with no expertise simply compares two generated motions. This is particularly efficient because evaluating the model's output is easier than gathering the motion that performs a desired task (e.g. backflip). To pioneer the exploration of this paradigm, we annotate 3,528 preference pairs generated by MotionGPT, marking the first effort to investigate various algorithms for learning from preference data. In particular, our exploration highlights important design choices when using preference data. Additionally, our experimental results show that preference learning has the potential to greatly improve current text-to-motion generative models. Our code and dataset are publicly available at https://github.com/THU-LYJ-Lab/InstructMotion}{https://github.com/THU-LYJ-Lab/InstructMotion to further facilitate research in this area.

Qiang He, Wentian Qu, Jiajia Dai et al.

Accurate semantic segmentation of 3D dental models is essential for digital dentistry applications such as orthodontics and dental implants. However, due to complex tooth arrangements and similarities in shape among adjacent teeth, existing methods struggle with accurate segmentation, because they often focus on local geometry while neglecting global contextual information. To address this, we propose TCATSeg, a novel framework that combines local geometric features with global semantic context. We introduce a set of sparse yet physically meaningful superpoints to capture global semantic relationships and enhance segmentation accuracy. Additionally, we present a new dataset of 400 dental models, including pre-orthodontic samples, to evaluate the generalization of our method. Extensive experiments demonstrate that TCATSeg outperforms state-of-the-art approaches.

Huan Liu, Shusen Yang, Yuzhe Zhang et al.

EEG-based emotion recognition (EER) has gained significant attention due to its potential for understanding and analyzing human emotions. While recent advancements in deep learning techniques have substantially improved EER, the field lacks a convincing benchmark and comprehensive open-source libraries. This absence complicates fair comparisons between models and creates reproducibility challenges for practitioners, which collectively hinder progress. To address these issues, we introduce LibEER, a comprehensive benchmark and algorithm library designed to facilitate fair comparisons in EER. LibEER carefully selects popular and powerful baselines, harmonizes key implementation details across methods, and provides a standardized codebase in PyTorch. By offering a consistent evaluation framework with standardized experimental settings, LibEER enables unbiased assessments of seventeen representative deep learning models for EER across the six most widely used datasets. Additionally, we conduct a thorough, reproducible comparison of model performance and efficiency, providing valuable insights to guide researchers in the selection and design of EER models. Moreover, we make observations and in-depth analysis on the experiment results and identify current challenges in this community. We hope that our work will not only lower entry barriers for newcomers to EEG-based emotion recognition but also contribute to the standardization of research in this domain, fostering steady development. The library and source code are publicly available at https://github.com/XJTU-EEG/LibEER.

Baiqiao Zhang, Xiangxian Li, Chao Zhou et al.

The low-intrusion and automated personality assessment is receiving increasing attention in psychology and human-computer interaction fields. This study explores an interactive approach for personality assessment, focusing on the multiplicity of personality representation. We propose a framework of Gamified Personality Assessment through Multi-Personality Representations (Multi-PR GPA). The framework leverages Large Language Models to empower virtual agents with different personalities. These agents elicit multifaceted human personality representations through engaging in interactive games. Drawing upon the multi-type textual data generated throughout the interaction, it achieves personality assessments with interpretable insights. Grounded in the classic Big Five personality theory, we developed a prototype system and conducted a user study to evaluate the efficacy of Multi-PR GPA. The results affirm the effectiveness of our approach in personality assessment and demonstrate its superior performance when considering the multiplicity of personality representation. Error structure analysis further revealed systematic assessment biases in LLMs, which multi-context aggregation partially mitigated.

Kaizhe Zhang, Yijie Zhou, Weizhan Zhang et al.

Dynamic novel view synthesis (NVS) is essential for creating immersive experiences. Existing approaches have advanced dynamic NVS by introducing 3D Gaussian Splatting (3DGS) with implicit deformation fields or indiscriminately assigned time-varying parameters, surpassing NeRF-based methods. However, due to excessive model complexity and parameter redundancy, they incur large model sizes and slow rendering speeds, making them inefficient for real-time applications, particularly on resource-constrained devices. To obtain a more efficient model with fewer redundant parameters, in this paper, we propose Hybrid Gaussian Splatting (HGS), a compact and efficient framework explicitly designed to disentangle static and dynamic regions of a scene within a unified representation. The core innovation of HGS lies in our Static-Dynamic Decomposition (SDD) strategy, which leverages Radial Basis Function (RBF) modeling for Gaussian primitives. Specifically, for dynamic regions, we employ time-dependent RBFs to effectively capture temporal variations and handle abrupt scene changes, while for static regions, we reduce redundancy by sharing temporally invariant parameters. Additionally, we introduce a two-stage training strategy tailored for explicit models to enhance temporal coherence at static-dynamic boundaries. Experimental results demonstrate that our method reduces model size by up to 98% and achieves real-time rendering at up to 125 FPS at 4K resolution on a single RTX 3090 GPU. It further sustains 160 FPS at 1352 * 1014 on an RTX 3050 and has been integrated into the VR system. Moreover, HGS achieves comparable rendering quality to state-of-the-art methods while providing significantly improved visual fidelity for high-frequency details and abrupt scene changes.

Huankun Sheng, Ming Li, Yixiang Wei et al.

Recent advances in object-centric representation learning have shown that slot attention-based methods can effectively decompose visual scenes into object slot representations without supervision. However, existing approaches typically process foreground and background regions indiscriminately, often resulting in background interference and suboptimal instance discovery performance on real-world data. To address this limitation, we propose Foreground-Aware Slot Attention (FASA), a two-stage framework that explicitly separates foreground from background to enable precise object discovery. In the first stage, FASA performs a coarse scene decomposition to distinguish foreground from background regions through a dual-slot competition mechanism. These slots are initialized via a clustering-based strategy, yielding well-structured representations of salient regions. In the second stage, we introduce a masked slot attention mechanism where the first slot captures the background while the remaining slots compete to represent individual foreground objects. To further address over-segmentation of foreground objects, we incorporate pseudo-mask guidance derived from a patch affinity graph constructed with self-supervised image features to guide the learning of foreground slots. Extensive experiments on both synthetic and real-world datasets demonstrate that FASA consistently outperforms state-of-the-art methods, validating the effectiveness of explicit foreground modeling and pseudo-mask guidance for robust scene decomposition and object-coherent representation. Code will be made publicly available.

Yubin Hu, Xiaoyang Guo, Yang Xiao et al.

This paper presents NGP-RT, a novel approach for enhancing the rendering speed of Instant-NGP to achieve real-time novel view synthesis. As a classic NeRF-based method, Instant-NGP stores implicit features in multi-level grids or hash tables and applies a shallow MLP to convert the implicit features into explicit colors and densities. Although it achieves fast training speed, there is still a lot of room for improvement in its rendering speed due to the per-point MLP executions for implicit multi-level feature aggregation, especially for real-time applications. To address this challenge, our proposed NGP-RT explicitly stores colors and densities as hash features, and leverages a lightweight attention mechanism to disambiguate the hash collisions instead of using computationally intensive MLP. At the rendering stage, NGP-RT incorporates a pre-computed occupancy distance grid into the ray marching strategy to inform the distance to the nearest occupied voxel, thereby reducing the number of marching points and global memory access. Experimental results show that on the challenging Mip-NeRF360 dataset, NGP-RT achieves better rendering quality than previous NeRF-based methods, achieving 108 fps at 1080p resolution on a single Nvidia RTX 3090 GPU. Our approach is promising for NeRF-based real-time applications that require efficient and high-quality rendering.

Sheng Ye, Zhen-Hui Dong, Ruoyu Fan et al.

Semantic understanding of 3D scenes is essential for robots to operate effectively and safely in complex environments. Existing methods for semantic scene reconstruction and semantic-aware novel view synthesis often rely on dense multi-view inputs and require scene-specific optimization, limiting their practicality and scalability in real-world applications. To address these challenges, we propose SemGS, a feed-forward framework for reconstructing generalizable semantic fields from sparse image inputs. SemGS uses a dual-branch architecture to extract color and semantic features, where the two branches share shallow CNN layers, allowing semantic reasoning to leverage textural and structural cues in color appearance. We also incorporate a camera-aware attention mechanism into the feature extractor to explicitly model geometric relationships between camera viewpoints. The extracted features are decoded into dual-Gaussians that share geometric consistency while preserving branch-specific attributes, and further rasterized to synthesize semantic maps under novel viewpoints. Additionally, we introduce a regional smoothness loss to enhance semantic coherence. Experiments show that SemGS achieves state-of-the-art performance on benchmark datasets, while providing rapid inference and strong generalization capabilities across diverse synthetic and real-world scenarios.

Zhiyao Sun, Ziqiao Peng, Yifeng Ma et al.

Real-time, streaming interactive avatars represent a critical yet challenging goal in digital human research. Although diffusion-based human avatar generation methods achieve remarkable success, their non-causal architecture and high computational costs make them unsuitable for streaming. Moreover, existing interactive approaches are typically limited to head-and-shoulder region, limiting their ability to produce gestures and body motions. To address these challenges, we propose a two-stage autoregressive adaptation and acceleration framework that applies autoregressive distillation and adversarial refinement to adapt a high-fidelity human video diffusion model for real-time, interactive streaming. To ensure long-term stability and consistency, we introduce three key components: a Reference Sink, a Reference-Anchored Positional Re-encoding (RAPR) strategy, and a Consistency-Aware Discriminator. Building on this framework, we develop a one-shot, interactive, human avatar model capable of generating both natural talking and listening behaviors with coherent gestures. Extensive experiments demonstrate that our method achieves state-of-the-art performance, surpassing existing approaches in generation quality, real-time efficiency, and interaction naturalness. Project page: https://streamavatar.github.io .

Changsong Lei, Yaqian Liang, Shaofeng Wang et al.

Digital orthodontics represents a prominent and critical application of computer vision technology in the medical field. So far, the labor-intensive process of collecting clinical data, particularly in acquiring paired 3D orthodontic teeth models, constitutes a crucial bottleneck for developing tooth arrangement neural networks. Although numerous general 3D shape generation methods have been proposed, most of them focus on single-object generation and are insufficient for generating anatomically structured teeth models, each comprising 24-32 segmented teeth. In this paper, we propose TeethGenerator, a novel two-stage framework designed to synthesize paired 3D teeth models pre- and post-orthodontic, aiming to facilitate the training of downstream tooth arrangement networks. Specifically, our approach consists of two key modules: (1) a teeth shape generation module that leverages a diffusion model to learn the distribution of morphological characteristics of teeth, enabling the generation of diverse post-orthodontic teeth models; and (2) a teeth style generation module that synthesizes corresponding pre-orthodontic teeth models by incorporating desired styles as conditional inputs. Extensive qualitative and quantitative experiments demonstrate that our synthetic dataset aligns closely with the distribution of real orthodontic data, and promotes tooth alignment performance significantly when combined with real data for training. The code and dataset are available at https://github.com/lcshhh/teeth_generator.

Ming Li, Yong-Jin Liu, Fang Liu et al.

Emotion recognition from multi-modal physiological and behavioral signals plays a pivotal role in affective computing, yet most existing models remain constrained to the prediction of singular emotions in controlled laboratory settings. Real-world human emotional experiences, by contrast, are often characterized by the simultaneous presence of multiple affective states, spurring recent interest in mixed emotion recognition as an emotion distribution learning problem. Current approaches, however, often neglect the valence consistency and structured correlations inherent among coexisting emotions. To address this limitation, we propose a Memory-guided Prototypical Co-occurrence Learning (MPCL) framework that explicitly models emotion co-occurrence patterns. Specifically, we first fuse multi-modal signals via a multi-scale associative memory mechanism. To capture cross-modal semantic relationships, we construct emotion-specific prototype memory banks, yielding rich physiological and behavioral representations, and employ prototype relation distillation to ensure cross-modal alignment in the latent prototype space. Furthermore, inspired by human cognitive memory systems, we introduce a memory retrieval strategy to extract semantic-level co-occurrence associations across emotion categories. Through this bottom-up hierarchical abstraction process, our model learns affectively informative representations for accurate emotion distribution prediction. Comprehensive experiments on two public datasets demonstrate that MPCL consistently outperforms state-of-the-art methods in mixed emotion recognition, both quantitatively and qualitatively.

Yuze He, Yanning Zhou, Wang Zhao et al.

We present StdGEN++, a novel and comprehensive system for generating high-fidelity, semantically decomposed 3D characters from diverse inputs. Existing 3D generative methods often produce monolithic meshes that lack the structural flexibility required by industrial pipelines in gaming and animation. Addressing this gap, StdGEN++ is built upon a Dual-branch Semantic-aware Large Reconstruction Model (Dual-Branch S-LRM), which jointly reconstructs geometry, color, and per-component semantics in a feed-forward manner. To achieve production-level fidelity, we introduce a novel semantic surface extraction formalism compatible with hybrid implicit fields. This mechanism is accelerated by a coarse-to-fine proposal scheme, which significantly reduces memory footprint and enables high-resolution mesh generation. Furthermore, we propose a video-diffusion-based texture decomposition module that disentangles appearance into editable layers (e.g., separated iris and skin), resolving semantic confusion in facial regions. Experiments demonstrate that StdGEN++ achieves state-of-the-art performance, significantly outperforming existing methods in geometric accuracy and semantic disentanglement. Crucially, the resulting structural independence unlocks advanced downstream capabilities, including non-destructive editing, physics-compliant animation, and gaze tracking, making it a robust solution for automated character asset production.

Wang Zhao, Jiachen Liu, Sheng Zhang et al.

This paper presents a generalizable 3D plane detection and reconstruction framework named MonoPlane. Unlike previous robust estimator-based works (which require multiple images or RGB-D input) and learning-based works (which suffer from domain shift), MonoPlane combines the best of two worlds and establishes a plane reconstruction pipeline based on monocular geometric cues, resulting in accurate, robust and scalable 3D plane detection and reconstruction in the wild. Specifically, we first leverage large-scale pre-trained neural networks to obtain the depth and surface normals from a single image. These monocular geometric cues are then incorporated into a proximity-guided RANSAC framework to sequentially fit each plane instance. We exploit effective 3D point proximity and model such proximity via a graph within RANSAC to guide the plane fitting from noisy monocular depths, followed by image-level multi-plane joint optimization to improve the consistency among all plane instances. We further design a simple but effective pipeline to extend this single-view solution to sparse-view 3D plane reconstruction. Extensive experiments on a list of datasets demonstrate our superior zero-shot generalizability over baselines, achieving state-of-the-art plane reconstruction performance in a transferring setting. Our code is available at https://github.com/thuzhaowang/MonoPlane .

Wang Zhao, Shaohui Liu, Yi Wei et al.

In this paper, we introduce a deep multi-view stereo (MVS) system that jointly predicts depths, surface normals and per-view confidence maps. The key to our approach is a novel solver that iteratively solves for per-view depth map and normal map by optimizing an energy potential based on the locally planar assumption. Specifically, the algorithm updates depth map by propagating from neighboring pixels with slanted planes, and updates normal map with local probabilistic plane fitting. Both two steps are monitored by a customized confidence map. This solver is not only effective as a post-processing tool for plane-based depth refinement and completion, but also differentiable such that it can be efficiently integrated into deep learning pipelines. Our multi-view stereo system employs multiple optimization steps of the solver over the initial prediction of depths and surface normals. The whole system can be trained end-to-end, decoupling the challenging problem of matching pixels within poorly textured regions from the cost-volume based neural network. Experimental results on ScanNet and RGB-D Scenes V2 demonstrate state-of-the-art performance of the proposed deep MVS system on multi-view depth estimation, with our proposed solver consistently improving the depth quality over both conventional and deep learning based MVS pipelines. Code is available at https://github.com/thuzhaowang/idn-solver.

Aihua Mao, Zihui Du, Junhui Hou et al.

Point cloud upsampling aims to generate dense point clouds from given sparse ones, which is a challenging task due to the irregular and unordered nature of point sets. To address this issue, we present a novel deep learning-based model, called PU-Flow, which incorporates normalizing flows and weight prediction techniques to produce dense points uniformly distributed on the underlying surface. Specifically, we exploit the invertible characteristics of normalizing flows to transform points between Euclidean and latent spaces and formulate the upsampling process as ensemble of neighbouring points in a latent space, where the ensemble weights are adaptively learned from local geometric context. Extensive experiments show that our method is competitive and, in most test cases, it outperforms state-of-the-art methods in terms of reconstruction quality, proximity-to-surface accuracy, and computation efficiency. The source code will be publicly available at https://github.com/unknownue/pu-flow.

Yudong Guo, Keyu Chen, Sen Liang et al.

Generating high-fidelity talking head video by fitting with the input audio sequence is a challenging problem that receives considerable attentions recently. In this paper, we address this problem with the aid of neural scene representation networks. Our method is completely different from existing methods that rely on intermediate representations like 2D landmarks or 3D face models to bridge the gap between audio input and video output. Specifically, the feature of input audio signal is directly fed into a conditional implicit function to generate a dynamic neural radiance field, from which a high-fidelity talking-head video corresponding to the audio signal is synthesized using volume rendering. Another advantage of our framework is that not only the head (with hair) region is synthesized as previous methods did, but also the upper body is generated via two individual neural radiance fields. Experimental results demonstrate that our novel framework can (1) produce high-fidelity and natural results, and (2) support free adjustment of audio signals, viewing directions, and background images. Code is available at https://github.com/YudongGuo/AD-NeRF.

Wang Zhao, Shaohui Liu, Yezhi Shu et al.

In this work, we tackle the essential problem of scale inconsistency for self-supervised joint depth-pose learning. Most existing methods assume that a consistent scale of depth and pose can be learned across all input samples, which makes the learning problem harder, resulting in degraded performance and limited generalization in indoor environments and long-sequence visual odometry application. To address this issue, we propose a novel system that explicitly disentangles scale from the network estimation. Instead of relying on PoseNet architecture, our method recovers relative pose by directly solving fundamental matrix from dense optical flow correspondence and makes use of a two-view triangulation module to recover an up-to-scale 3D structure. Then, we align the scale of the depth prediction with the triangulated point cloud and use the transformed depth map for depth error computation and dense reprojection check. Our whole system can be jointly trained end-to-end. Extensive experiments show that our system not only reaches state-of-the-art performance on KITTI depth and flow estimation, but also significantly improves the generalization ability of existing self-supervised depth-pose learning methods under a variety of challenging scenarios, and achieves state-of-the-art results among self-supervised learning-based methods on KITTI Odometry and NYUv2 dataset. Furthermore, we present some interesting findings on the limitation of PoseNet-based relative pose estimation methods in terms of generalization ability. Code is available at https://github.com/B1ueber2y/TrianFlow.

Aihua Mao, Kaihang Huang, Yong-Jin Liu et al.

3D object affordance grounding aims to identify regions on 3D objects that support human-object interaction (HOI), a capability essential to embodied visual reasoning. However, most existing approaches rely on static visual or textual cues, neglecting that affordances are inherently defined by dynamic actions. As a result, they often struggle to localize the true contact regions involved in real interactions. We take a different perspective. Humans learn how to use objects by observing and imitating actions, not just by examining shapes. Motivated by this intuition, we introduce video-guided 3D affordance grounding, which leverages dynamic interaction sequences to provide functional supervision. To achieve this, we propose VAGNet, a framework that aligns video-derived interaction cues with 3D structure to resolve ambiguities that static cues cannot address. To support this new setting, we introduce PVAD, the first HOI video-3D pairing affordance dataset, providing functional supervision unavailable in prior works. Extensive experiments on PVAD show that VAGNet achieves state-of-the-art performance, significantly outperforming static-based baselines. The code and dataset will be open publicly.

Mengfei Xia, Yu Zhou, Ran Yi et al.

Applying diffusion models to image-to-image translation (I2I) has recently received increasing attention due to its practical applications. Previous attempts inject information from the source image into each denoising step for an iterative refinement, thus resulting in a time-consuming implementation. We propose an efficient method that equips a diffusion model with a lightweight translator, dubbed a Diffusion Model Translator (DMT), to accomplish I2I. Specifically, we first offer theoretical justification that in employing the pioneering DDPM work for the I2I task, it is both feasible and sufficient to transfer the distribution from one domain to another only at some intermediate step. We further observe that the translation performance highly depends on the chosen timestep for domain transfer, and therefore propose a practical strategy to automatically select an appropriate timestep for a given task. We evaluate our approach on a range of I2I applications, including image stylization, image colorization, segmentation to image, and sketch to image, to validate its efficacy and general utility. The comparisons show that our DMT surpasses existing methods in both quality and efficiency. Code will be made publicly available.

Yuze He, Yanning Zhou, Wang Zhao et al.

We present StdGEN, an innovative pipeline for generating semantically decomposed high-quality 3D characters from single images, enabling broad applications in virtual reality, gaming, and filmmaking, etc. Unlike previous methods which struggle with limited decomposability, unsatisfactory quality, and long optimization times, StdGEN features decomposability, effectiveness and efficiency; i.e., it generates intricately detailed 3D characters with separated semantic components such as the body, clothes, and hair, in three minutes. At the core of StdGEN is our proposed Semantic-aware Large Reconstruction Model (S-LRM), a transformer-based generalizable model that jointly reconstructs geometry, color and semantics from multi-view images in a feed-forward manner. A differentiable multi-layer semantic surface extraction scheme is introduced to acquire meshes from hybrid implicit fields reconstructed by our S-LRM. Additionally, a specialized efficient multi-view diffusion model and an iterative multi-layer surface refinement module are integrated into the pipeline to facilitate high-quality, decomposable 3D character generation. Extensive experiments demonstrate our state-of-the-art performance in 3D anime character generation, surpassing existing baselines by a significant margin in geometry, texture and decomposability. StdGEN offers ready-to-use semantic-decomposed 3D characters and enables flexible customization for a wide range of applications. Project page: https://stdgen.github.io

Yuze He, Wang Zhao, Shaohui Liu et al.

We introduce AlphaTablets, a novel and generic representation of 3D planes that features continuous 3D surface and precise boundary delineation. By representing 3D planes as rectangles with alpha channels, AlphaTablets combine the advantages of current 2D and 3D plane representations, enabling accurate, consistent and flexible modeling of 3D planes. We derive differentiable rasterization on top of AlphaTablets to efficiently render 3D planes into images, and propose a novel bottom-up pipeline for 3D planar reconstruction from monocular videos. Starting with 2D superpixels and geometric cues from pre-trained models, we initialize 3D planes as AlphaTablets and optimize them via differentiable rendering. An effective merging scheme is introduced to facilitate the growth and refinement of AlphaTablets. Through iterative optimization and merging, we reconstruct complete and accurate 3D planes with solid surfaces and clear boundaries. Extensive experiments on the ScanNet dataset demonstrate state-of-the-art performance in 3D planar reconstruction, underscoring the great potential of AlphaTablets as a generic 3D plane representation for various applications. Project page is available at: https://hyzcluster.github.io/alphatablets

Jingwen Ye, Yuze He, Yanning Zhou et al.

Shape primitive abstraction, which decomposes complex 3D shapes into simple geometric elements, plays a crucial role in human visual cognition and has broad applications in computer vision and graphics. While recent advances in 3D content generation have shown remarkable progress, existing primitive abstraction methods either rely on geometric optimization with limited semantic understanding or learn from small-scale, category-specific datasets, struggling to generalize across diverse shape categories. We present PrimitiveAnything, a novel framework that reformulates shape primitive abstraction as a primitive assembly generation task. PrimitiveAnything includes a shape-conditioned primitive transformer for auto-regressive generation and an ambiguity-free parameterization scheme to represent multiple types of primitives in a unified manner. The proposed framework directly learns the process of primitive assembly from large-scale human-crafted abstractions, enabling it to capture how humans decompose complex shapes into primitive elements. Through extensive experiments, we demonstrate that PrimitiveAnything can generate high-quality primitive assemblies that better align with human perception while maintaining geometric fidelity across diverse shape categories. It benefits various 3D applications and shows potential for enabling primitive-based user-generated content (UGC) in games. Project page: https://primitiveanything.github.io

Xianhe Jiao, Chenlei Lv, Junli Zhao et al.

For large-scale point cloud processing, resampling takes the important role of controlling the point number and density while keeping the geometric consistency. % in related tasks. However, current methods cannot balance such different requirements. Particularly with large-scale point clouds, classical methods often struggle with decreased efficiency and accuracy. To address such issues, we propose a weighted Poisson-disk (WPD) resampling method to improve the usability and efficiency for the processing. We first design an initial Poisson resampling with a voxel-based estimation strategy. It is able to estimate a more accurate radius of the Poisson-disk while maintaining high efficiency. Then, we design a weighted tangent smoothing step to further optimize the Voronoi diagram for each point. At the same time, sharp features are detected and kept in the optimized results with isotropic property. Finally, we achieve a resampling copy from the original point cloud with the specified point number, uniform density, and high-quality geometric consistency. Experiments show that our method significantly improves the performance of large-scale point cloud resampling for different applications, and provides a highly practical solution.

Yubin Hu, Kairui Wen, Heng Zhou et al.

Reconstructing accurate 3D surfaces for street-view scenarios is crucial for applications such as digital entertainment and autonomous driving simulation. However, existing street-view datasets, including KITTI, Waymo, and nuScenes, only offer noisy LiDAR points as ground-truth data for geometric evaluation of reconstructed surfaces. These geometric ground-truths often lack the necessary precision to evaluate surface positions and do not provide data for assessing surface normals. To overcome these challenges, we introduce the SS3DM dataset, comprising precise \textbf{S}ynthetic \textbf{S}treet-view \textbf{3D} \textbf{M}esh models exported from the CARLA simulator. These mesh models facilitate accurate position evaluation and include normal vectors for evaluating surface normal. To simulate the input data in realistic driving scenarios for 3D reconstruction, we virtually drive a vehicle equipped with six RGB cameras and five LiDAR sensors in diverse outdoor scenes. Leveraging this dataset, we establish a benchmark for state-of-the-art surface reconstruction methods, providing a comprehensive evaluation of the associated challenges. For more information, visit our homepage at https://ss3dm.top.

Matthieu Lin, Jenny Sheng, Andrew Zhao et al. · tsinghua

This survey organizes the intricate literature on the design and optimization of emerging structures around post-trained LMs. We refer to this overarching structure as scaffolded LMs and focus on LMs that are integrated into multi-step processes with tools. We view scaffolded LMs as semi-parametric models wherein we train non-parametric variables, including the prompt, tools, and scaffold's code. In particular, they interpret instructions, use tools, and receive feedback all in language. Recent works use an LM as an optimizer to interpret language supervision and update non-parametric variables according to intricate objectives. In this survey, we refer to this paradigm as training of scaffolded LMs with language supervision. A key feature of non-parametric training is the ability to learn from language. Parametric training excels in learning from demonstration (supervised learning), exploration (reinforcement learning), or observations (unsupervised learning), using well-defined loss functions. Language-based optimization enables rich, interpretable, and expressive objectives, while mitigating issues like catastrophic forgetting and supporting compatibility with closed-source models. Furthermore, agents are increasingly deployed as co-workers in real-world applications such as Copilot in Office tools or software development. In these mixed-autonomy settings, where control and decision-making are shared between human and AI, users point out errors or suggest corrections. Accordingly, we discuss agents that continuously improve by learning from this real-time, language-based feedback and refer to this setting as streaming learning from language supervision.

Yuze He, Yushi Bai, Matthieu Lin et al. · tsinghua

By lifting the pre-trained 2D diffusion models into Neural Radiance Fields (NeRFs), text-to-3D generation methods have made great progress. Many state-of-the-art approaches usually apply score distillation sampling (SDS) to optimize the NeRF representations, which supervises the NeRF optimization with pre-trained text-conditioned 2D diffusion models such as Imagen. However, the supervision signal provided by such pre-trained diffusion models only depends on text prompts and does not constrain the multi-view consistency. To inject the cross-view consistency into diffusion priors, some recent works finetune the 2D diffusion model with multi-view data, but still lack fine-grained view coherence. To tackle this challenge, we incorporate multi-view image conditions into the supervision signal of NeRF optimization, which explicitly enforces fine-grained view consistency. With such stronger supervision, our proposed text-to-3D method effectively mitigates the generation of floaters (due to excessive densities) and completely empty spaces (due to insufficient densities). Our quantitative evaluations on the T$^3$Bench dataset demonstrate that our method achieves state-of-the-art performance over existing text-to-3D methods. We will make the code publicly available.

Wen Wen, Tieliang Gong, Yunjiao Zhang et al.

Continual learning (CL) has emerged as a dominant paradigm for acquiring knowledge from sequential tasks while avoiding catastrophic forgetting. Although many CL methods have been proposed to show impressive empirical performance, the theoretical understanding of their generalization behavior remains limited, particularly for replay-based approaches. In this paper, we establish a unified theoretical framework for replay-based CL, deriving a series of information-theoretic bounds that explicitly characterize how the memory buffer interacts with the current task to affect generalization. Specifically, our hypothesis-based bounds reveal that utilizing the limited exemplars of previous tasks alongside the current task data, rather than exhaustive replay, facilitates improved generalization while effectively mitigating catastrophic forgetting. Furthermore, our prediction-based bounds yield tighter and computationally tractable upper bounds of the generalization gap through the use of low-dimensional variables. Our analysis is general and broadly applicable to a wide range of learning algorithms, exemplified by stochastic gradient Langevin dynamics (SGLD) as a representative method. Comprehensive experimental evaluations demonstrate the effectiveness of our derived bounds in capturing the generalization dynamics in replay-based CL settings.

Sheng Fang, Yong-Jin Liu, Wei Yao et al.

Bilevel optimization, addressing challenges in hierarchical learning tasks, has gained significant interest in machine learning. The practical implementation of the gradient descent method to bilevel optimization encounters computational hurdles, notably the computation of the exact lower-level solution and the inverse Hessian of the lower-level objective. Although these two aspects are inherently connected, existing methods typically handle them separately by solving the lower-level problem and a linear system for the inverse Hessian-vector product. In this paper, we introduce a general framework to address these computational challenges in a coordinated manner. Specifically, we leverage quasi-Newton algorithms to accelerate the resolution of the lower-level problem while efficiently approximating the inverse Hessian-vector product. Furthermore, by exploiting the superlinear convergence properties of BFGS, we establish the non-asymptotic convergence analysis of the BFGS adaptation within our framework. Numerical experiments demonstrate the comparable or superior performance of the proposed algorithms in real-world learning tasks, including hyperparameter optimization, data hyper-cleaning, and few-shot meta-learning.

Wenqi Ji, Fang liu, Xinxin Du et al.

Interpersonal relationship quality is pivotal in social and occupational contexts. Existing analysis of interpersonal relationships mostly rely on subjective self-reports, whereas objective quantification remains challenging. In this paper, we propose a novel social relationship analysis framework using spatio-temporal patterns derived from dyadic EEG signals, which can be applied to quantitatively measure team cooperation in corporate team building, and evaluate interpersonal dynamics between therapists and patients in psychiatric therapy. First, we constructed a dyadic-EEG dataset from 72 pairs of participants with two relationships (stranger or friend) when watching emotional videos simultaneously. Then we proposed a deep neural network on dyadic-subject EEG signals, in which we combine the dynamic graph convolutional neural network for characterizing the interpersonal relationships among the EEG channels and 1-dimension convolution for extracting the information from the time sequence. To obtain the feature vectors from two EEG recordings that well represent the relationship of two subjects, we integrate deep canonical correlation analysis and triplet loss for training the network. Experimental results show that the social relationship type (stranger or friend) between two individuals can be effectively identified through their EEG data.

Wenhao Yan, Sheng Ye, Zhuoyi Yang et al.

Achieving character animation that meets studio-grade production standards remains challenging despite recent progress. Existing approaches can transfer motion from a driving video to a reference image, but often fail to preserve structural fidelity and temporal consistency in wild scenarios involving complex motion and cross-identity animations. In this work, we present \textbf{SCAIL} (a framework toward \textbf{S}tudio-grade \textbf{C}haracter \textbf{A}nimation via \textbf{I}n-context \textbf{L}earning), a framework designed to address these challenges from two key innovations. First, we propose a novel 3D pose representation, providing a more robust and flexible motion signal. Second, we introduce a full-context pose injection mechanism within a diffusion-transformer architecture, enabling effective spatio-temporal reasoning over full motion sequences. To align with studio-level requirements, we develop a curated data pipeline ensuring both diversity and quality, and establish a comprehensive benchmark for systematic evaluation. Experiments show that \textbf{SCAIL} achieves state-of-the-art performance and advances character animation toward studio-grade reliability and realism.