Fei Deng

Aiyuan Yang, Bin Xiao, Bingning Wang et al. · pku

Large language models (LLMs) have demonstrated remarkable performance on a variety of natural language tasks based on just a few examples of natural language instructions, reducing the need for extensive feature engineering. However, most powerful LLMs are closed-source or limited in their capability for languages other than English. In this technical report, we present Baichuan 2, a series of large-scale multilingual language models containing 7 billion and 13 billion parameters, trained from scratch, on 2.6 trillion tokens. Baichuan 2 matches or outperforms other open-source models of similar size on public benchmarks like MMLU, CMMLU, GSM8K, and HumanEval. Furthermore, Baichuan 2 excels in vertical domains such as medicine and law. We will release all pre-training model checkpoints to benefit the research community in better understanding the training dynamics of Baichuan 2.

Fei Deng, Yanwu Xu, Zhipeng Bao et al.

The remarkable generation quality of modern diffusion models often comes at the cost of massive parameter counts, which necessitate server-side inference with significant computational costs and potential privacy risks. Consequently, there is growing momentum toward developing efficient on-device alternatives. While recent efforts have optimized text-to-image models for mobile hardware, they remain relatively bulky, typically ranging from 0.5B to 1B parameters. We present BlazeEdit, a highly efficient, generalist image-to-image diffusion model tailored for on-device deployment. By identifying that many practical image editing tasks do not require text-based guidance, we eliminate the text-conditioning components and develop a multi-task architecture that consolidates object removal, outpainting, tone correction, relighting, and sticker generation into a single, compact model of only 195M parameters. BlazeEdit achieves a substantial reduction in download size and memory overhead while maintaining competitive generation quality. It completes a full inference pass in just 290ms on a Pixel 10, delivering a seamless, privacy-preserving, and lightning-fast experience for generalist image editing on the edge.

Ligong Han, Seungwook Han, Shivchander Sudalairaj et al.

Transformations based on domain expertise (expert transformations), such as random-resized-crop and color-jitter, have proven critical to the success of contrastive learning techniques such as SimCLR. Recently, several attempts have been made to replace such domain-specific, human-designed transformations with generated views that are learned. However for imagery data, so far none of these view-generation methods has been able to outperform expert transformations. In this work, we tackle a different question: instead of replacing expert transformations with generated views, can we constructively assimilate generated views with expert transformations? We answer this question in the affirmative and propose a view generation method and a simple, effective assimilation method that together improve the state-of-the-art by up to ~3.6% on three different datasets. Importantly, we conduct a detailed empirical study that systematically analyzes a range of view generation and assimilation methods and provides a holistic picture of the efficacy of learned views in contrastive representation learning.

Qingsong Yan, Qiang Wang, Kaiyong Zhao et al.

Existing learning-based multi-view stereo (MVS) methods rely on the depth range to build the 3D cost volume and may fail when the range is too large or unreliable. To address this problem, we propose a disparity-based MVS method based on the epipolar disparity flow (E-flow), called DispMVS, which infers the depth information from the pixel movement between two views. The core of DispMVS is to construct a 2D cost volume on the image plane along the epipolar line between each pair (between the reference image and several source images) for pixel matching and fuse uncountable depths triangulated from each pair by multi-view geometry to ensure multi-view consistency. To be robust, DispMVS starts from a randomly initialized depth map and iteratively refines the depth map with the help of the coarse-to-fine strategy. Experiments on DTUMVS and Tanks\&Temple datasets show that DispMVS is not sensitive to the depth range and achieves state-of-the-art results with lower GPU memory.

Fei Deng, Junyeong Park, Sungjin Ahn

World models are a fundamental component in model-based reinforcement learning (MBRL). To perform temporally extended and consistent simulations of the future in partially observable environments, world models need to possess long-term memory. However, state-of-the-art MBRL agents, such as Dreamer, predominantly employ recurrent neural networks (RNNs) as their world model backbone, which have limited memory capacity. In this paper, we seek to explore alternative world model backbones for improving long-term memory. In particular, we investigate the effectiveness of Transformers and Structured State Space Sequence (S4) models, motivated by their remarkable ability to capture long-range dependencies in low-dimensional sequences and their complementary strengths. We propose S4WM, the first world model compatible with parallelizable SSMs including S4 and its variants. By incorporating latent variable modeling, S4WM can efficiently generate high-dimensional image sequences through latent imagination. Furthermore, we extensively compare RNN-, Transformer-, and S4-based world models across four sets of environments, which we have tailored to assess crucial memory capabilities of world models, including long-term imagination, context-dependent recall, reward prediction, and memory-based reasoning. Our findings demonstrate that S4WM outperforms Transformer-based world models in terms of long-term memory, while exhibiting greater efficiency during training and imagination. These results pave the way for the development of stronger MBRL agents.

Qingsong Yan, Qiang Wang, Kaiyong Zhao et al.

The panorama image can simultaneously demonstrate complete information of the surrounding environment and has many advantages in virtual tourism, games, robotics, etc. However, the progress of panorama depth estimation cannot completely solve the problems of distortion and discontinuity caused by the commonly used projection methods. This paper proposes SphereDepth, a novel panorama depth estimation method that predicts the depth directly on the spherical mesh without projection preprocessing. The core idea is to establish the relationship between the panorama image and the spherical mesh and then use a deep neural network to extract features on the spherical domain to predict depth. To address the efficiency challenges brought by the high-resolution panorama data, we introduce two hyper-parameters for the proposed spherical mesh processing framework to balance the inference speed and accuracy. Validated on three public panorama datasets, SphereDepth achieves comparable results with the state-of-the-art methods of panorama depth estimation. Benefiting from the spherical domain setting, SphereDepth can generate a high-quality point cloud and significantly alleviate the issues of distortion and discontinuity.

Baichuan-M2 Team, Chengfeng Dou, Chong Liu et al.

As large language models (LLMs) advance in conversational and reasoning capabilities, their practical application in healthcare has become a critical research focus. However, there is a notable gap between the performance of medical LLMs on static benchmarks such as USMLE and their utility in real-world clinical decision-making. This discrepancy arises because traditional exams fail to capture the dynamic, interactive nature of medical consultations. To address this challenge, we introduce a novel dynamic verification framework that moves beyond static answer verifier, establishing a large-scale, high-fidelity interactive reinforcement learning system. Our framework comprises two key components: a Patient Simulator that creates realistic clinical environments using de-identified medical records, and a Clinical Rubrics Generator that dynamically produces multi-dimensional evaluation metrics. Building on this foundation, we develop Baichuan-M2, a 32B-parameter medical augmented reasoning model trained through a multi-stage reinforcement learning strategy with an improved Group Relative Policy Optimization (GRPO) algorithm. Evaluated on HealthBench, Baichuan-M2 outperforms all other open-source models and most advanced closed-source counterparts, achieving a score above 32 on the challenging HealthBench Hard benchmark-previously exceeded only by GPT-5. Our work demonstrates that robust dynamic verifier system is essential for aligning LLM capabilities with practical clinical applications, establishing a new Pareto front in the performance-parameter trade-off for medical AI deployment.

Shuang Wang, Fei Deng, Peifan Jiang et al.

Geographical, physical, or economic constraints often result in missing traces within seismic data, making the reconstruction of complete seismic data a crucial step in seismic data processing. Traditional methods for seismic data reconstruction require the selection of multiple empirical parameters and struggle to handle large-scale continuous missing data. With the development of deep learning, various neural networks have demonstrated powerful reconstruction capabilities. However, these convolutional neural networks represent a point-to-point reconstruction approach that may not cover the entire distribution of the dataset. Consequently, when dealing with seismic data featuring complex missing patterns, such networks may experience varying degrees of performance degradation. In response to this challenge, we propose a novel diffusion model reconstruction framework tailored for 3D seismic data. To constrain the results generated by the diffusion model, we introduce conditional supervision constraints into the diffusion model, constraining the generated data of the diffusion model based on the input data to be reconstructed. We introduce a 3D neural network architecture into the diffusion model, successfully extending the 2D diffusion model to 3D space. Additionally, we refine the model's generation process by incorporating missing data into the generation process, resulting in reconstructions with higher consistency. Through ablation studies determining optimal parameter values, our method exhibits superior reconstruction accuracy when applied to both field datasets and synthetic datasets, effectively addressing a wide range of complex missing patterns. Our implementation is available at https://github.com/WAL-l/SeisFusion.

Jindong Jiang, Fei Deng, Gautam Singh et al.

The recent success of transformer-based image generative models in object-centric learning highlights the importance of powerful image generators for handling complex scenes. However, despite the high expressiveness of diffusion models in image generation, their integration into object-centric learning remains largely unexplored in this domain. In this paper, we explore the feasibility and potential of integrating diffusion models into object-centric learning and investigate the pros and cons of this approach. We introduce Latent Slot Diffusion (LSD), a novel model that serves dual purposes: it is the first object-centric learning model to replace conventional slot decoders with a latent diffusion model conditioned on object slots, and it is also the first unsupervised compositional conditional diffusion model that operates without the need for supervised annotations like text. Through experiments on various object-centric tasks, including the first application of the FFHQ dataset in this field, we demonstrate that LSD significantly outperforms state-of-the-art transformer-based decoders, particularly in more complex scenes, and exhibits superior unsupervised compositional generation quality. In addition, we conduct a preliminary investigation into the integration of pre-trained diffusion models in LSD and demonstrate its effectiveness in real-world image segmentation and generation. Project page is available at https://latentslotdiffusion.github.io

Bin Wang, Fei Deng, Shuang Wang et al.

Semantic segmentation of remote sensing (RS) images is a challenging yet essential task with broad applications. While deep learning, particularly supervised learning with large-scale labeled datasets, has significantly advanced this field, the acquisition of high-quality labeled data remains costly and time-intensive. Unsupervised domain adaptation (UDA) provides a promising alternative by enabling models to learn from unlabeled target domain data while leveraging labeled source domain data. Recent self-training (ST) approaches employing pseudo-label generation have shown potential in mitigating domain discrepancies. However, the application of ST to RS image segmentation remains underexplored. Factors such as variations in ground sampling distance, imaging equipment, and geographic diversity exacerbate domain shifts, limiting model performance across domains. In that case, existing ST methods, due to significant domain shifts in cross-domain RS images, often underperform. To address these challenges, we propose integrating contrastive learning into UDA, enhancing the model's ability to capture semantic information in the target domain by maximizing the similarity between augmented views of the same image. This additional supervision improves the model's representational capacity and segmentation performance in the target domain. Extensive experiments conducted on RS datasets, including Potsdam, Vaihingen, and LoveDA, demonstrate that our method, SimSeg, outperforms existing approaches, achieving state-of-the-art results. Visualization and quantitative analyses further validate SimSeg's superior ability to learn from the target domain. The code is publicly available at https://github.com/woldier/SiamSeg.

Fei Deng, Ingook Jang, Sungjin Ahn

Top-performing Model-Based Reinforcement Learning (MBRL) agents, such as Dreamer, learn the world model by reconstructing the image observations. Hence, they often fail to discard task-irrelevant details and struggle to handle visual distractions. To address this issue, previous work has proposed to contrastively learn the world model, but the performance tends to be inferior in the absence of distractions. In this paper, we seek to enhance robustness to distractions for MBRL agents. Specifically, we consider incorporating prototypical representations, which have yielded more accurate and robust results than contrastive approaches in computer vision. However, it remains elusive how prototypical representations can benefit temporal dynamics learning in MBRL, since they treat each image independently without capturing temporal structures. To this end, we propose to learn the prototypes from the recurrent states of the world model, thereby distilling temporal structures from past observations and actions into the prototypes. The resulting model, DreamerPro, successfully combines Dreamer with prototypes, making large performance gains on the DeepMind Control suite both in the standard setting and when there are complex background distractions. Code available at https://github.com/fdeng18/dreamer-pro .

Fei Deng, Yinghui He, Chuntong Chu et al.

Human Activity Recognition (HAR) in smart homes is critical for health monitoring and assistive living. While vision-based systems are common, they face privacy concerns and environmental limitations (e.g., occlusion). In this work, we present MobiDiary, a framework that generates natural language descriptions of daily activities directly from heterogeneous physical signals (specifically IMU and Wi-Fi). Unlike conventional approaches that restrict outputs to pre-defined labels, MobiDiary produces expressive, human-readable summaries. To bridge the semantic gap between continuous, noisy physical signals and discrete linguistic descriptions, we propose a unified sensor encoder. Instead of relying on modality-specific engineering, we exploit the shared inductive biases of motion-induced signals--where both inertial and wireless data reflect underlying kinematic dynamics. Specifically, our encoder utilizes a patch-based mechanism to capture local temporal correlations and integrates heterogeneous placement embedding to unify spatial contexts across different sensors. These unified signal tokens are then fed into a Transformer-based decoder, which employs an autoregressive mechanism to generate coherent action descriptions word-by-word. We comprehensively evaluate our approach on multiple public benchmarks (XRF V2, UWash, and WiFiTAD). Experimental results demonstrate that MobiDiary effectively generalizes across modalities, achieving state-of-the-art performance on captioning metrics (e.g., BLEU@4, CIDEr, RMC) and outperforming specialized baselines in continuous action understanding.

Chang Chen, Fei Deng, Kenji Kawaguchi et al.

Diffusion-based generative methods have proven effective in modeling trajectories with offline datasets. However, they often face computational challenges and can falter in generalization, especially in capturing temporal abstractions for long-horizon tasks. To overcome this, we introduce the Hierarchical Diffuser, a simple, fast, yet surprisingly effective planning method combining the advantages of hierarchical and diffusion-based planning. Our model adopts a "jumpy" planning strategy at the higher level, which allows it to have a larger receptive field but at a lower computational cost -- a crucial factor for diffusion-based planning methods, as we have empirically verified. Additionally, the jumpy sub-goals guide our low-level planner, facilitating a fine-tuning stage and further improving our approach's effectiveness. We conducted empirical evaluations on standard offline reinforcement learning benchmarks, demonstrating our method's superior performance and efficiency in terms of training and planning speed compared to the non-hierarchical Diffuser as well as other hierarchical planning methods. Moreover, we explore our model's generalization capability, particularly on how our method improves generalization capabilities on compositional out-of-distribution tasks.

Fei Deng, Qifei Wang, Wei Wei et al.

Reward finetuning has emerged as a promising approach to aligning foundation models with downstream objectives. Remarkable success has been achieved in the language domain by using reinforcement learning (RL) to maximize rewards that reflect human preference. However, in the vision domain, existing RL-based reward finetuning methods are limited by their instability in large-scale training, rendering them incapable of generalizing to complex, unseen prompts. In this paper, we propose Proximal Reward Difference Prediction (PRDP), enabling stable black-box reward finetuning for diffusion models for the first time on large-scale prompt datasets with over 100K prompts. Our key innovation is the Reward Difference Prediction (RDP) objective that has the same optimal solution as the RL objective while enjoying better training stability. Specifically, the RDP objective is a supervised regression objective that tasks the diffusion model with predicting the reward difference of generated image pairs from their denoising trajectories. We theoretically prove that the diffusion model that obtains perfect reward difference prediction is exactly the maximizer of the RL objective. We further develop an online algorithm with proximal updates to stably optimize the RDP objective. In experiments, we demonstrate that PRDP can match the reward maximization ability of well-established RL-based methods in small-scale training. Furthermore, through large-scale training on text prompts from the Human Preference Dataset v2 and the Pick-a-Pic v1 dataset, PRDP achieves superior generation quality on a diverse set of complex, unseen prompts whereas RL-based methods completely fail.

Seung Hyun Lee, Yinxiao Li, Junjie Ke et al.

Recent works have demonstrated that using reinforcement learning (RL) with multiple quality rewards can improve the quality of generated images in text-to-image (T2I) generation. However, manually adjusting reward weights poses challenges and may cause over-optimization in certain metrics. To solve this, we propose Parrot, which addresses the issue through multi-objective optimization and introduces an effective multi-reward optimization strategy to approximate Pareto optimal. Utilizing batch-wise Pareto optimal selection, Parrot automatically identifies the optimal trade-off among different rewards. We use the novel multi-reward optimization algorithm to jointly optimize the T2I model and a prompt expansion network, resulting in significant improvement of image quality and also allow to control the trade-off of different rewards using a reward related prompt during inference. Furthermore, we introduce original prompt-centered guidance at inference time, ensuring fidelity to user input after prompt expansion. Extensive experiments and a user study validate the superiority of Parrot over several baselines across various quality criteria, including aesthetics, human preference, text-image alignment, and image sentiment.

Bin Wang, Fei Deng, Peifan Jiang

Electroencephalogram (EEG) signals play a pivotal role in clinical medicine, brain research, and neurological disease studies. However, susceptibility to various physiological and environmental artifacts introduces noise in recorded EEG data, impeding accurate analysis of underlying brain activity. Denoising techniques are crucial to mitigate this challenge. Recent advancements in deep learningbased approaches exhibit substantial potential for enhancing the signal-to-noise ratio of EEG data compared to traditional methods. In the realm of large-scale language models (LLMs), the Retentive Network (Retnet) infrastructure, prevalent for some models, demonstrates robust feature extraction and global modeling capabilities. Recognizing the temporal similarities between EEG signals and natural language, we introduce the Retnet from natural language processing to EEG denoising. This integration presents a novel approach to EEG denoising, opening avenues for a profound understanding of brain activities and accurate diagnosis of neurological diseases. Nonetheless, direct application of Retnet to EEG denoising is unfeasible due to the one-dimensional nature of EEG signals, while natural language processing deals with two-dimensional data. To facilitate Retnet application to EEG denoising, we propose the signal embedding method, transforming one-dimensional EEG signals into two dimensions for use as network inputs. Experimental results validate the substantial improvement in denoising effectiveness achieved by the proposed method.

Qingsong Yan, Qiang Wang, Kaiyong Zhao et al.

Neural Radiance Fields (NeRF) have demonstrated impressive performance in novel view synthesis. However, NeRF and most of its variants still rely on traditional complex pipelines to provide extrinsic and intrinsic camera parameters, such as COLMAP. Recent works, like NeRFmm, BARF, and L2G-NeRF, directly treat camera parameters as learnable and estimate them through differential volume rendering. However, these methods work for forward-looking scenes with slight motions and fail to tackle the rotation scenario in practice. To overcome this limitation, we propose a novel \underline{c}amera parameter \underline{f}ree neural radiance field (CF-NeRF), which incrementally reconstructs 3D representations and recovers the camera parameters inspired by incremental structure from motion (SfM). Given a sequence of images, CF-NeRF estimates the camera parameters of images one by one and reconstructs the scene through initialization, implicit localization, and implicit optimization. To evaluate our method, we use a challenging real-world dataset NeRFBuster which provides 12 scenes under complex trajectories. Results demonstrate that CF-NeRF is robust to camera rotation and achieves state-of-the-art results without providing prior information and constraints.

Shihui Yang, Chengfeng Dou, Peidong Guo et al.

Reinforcement Learning from Verifiable Rewards (RLVR) has emerged as a promising framework for enhancing the reasoning capabilities of large language models. However, existing approaches such as GRPO often suffer from zero gradients. This problem arises primarily due to fixed clipping bounds for token-level probability ratios and the standardization of identical rewards, which can lead to ineffective gradient updates and underutilization of generated responses. In this work, we propose Dynamic Clipping Policy Optimization(DCPO), which introduces a dynamic clipping strategy that adaptively adjusts clipping bounds based on token-specific prior probabilities to enhance token-level exploration, and a smooth advantage standardization technique that standardizes rewards across cumulative training steps to improve the response-level effective utilization of generated responses. DCPO achieved state-of-the-art performance on four benchmarks based on four different models. In particular, DCPO achieved an Avg@1 of 46.7 under greedy decoding and an Avg@32 of 38.8 under 32 times sampling on the AIME24 benchmark, surpassing DAPO (36.7/31.6), GRPO (36.7/32.1) and GSPO (40.0/34.9) on the Qwen2.5-Math-7B model. On the AIME25 benchmark based on Qwen2.5-14B, DCPO achieves a performance of (23.3/19.0), surpassing GRPO (13.3/10.5), DAPO (20.0/15.3) and GSPO (16.7/9.9). Furthermore, DCPO achieved an average 28% improvement in the nonzero advantage over GRPO in four models, doubled the training efficiency over DAPO, and significantly reduced the token clipping ratio by an order of magnitude compared to both GRPO and DAPO, while achieving superior performance. These results highlight DCPO's effectiveness in leveraging generated data more efficiently for reinforcement learning in large language models.

Bin Wang, Fei Deng, Peifan Jiang et al.

Low-dose computed tomography (LDCT) has become the technology of choice for diagnostic medical imaging, given its lower radiation dose compared to standard CT, despite increasing image noise and potentially affecting diagnostic accuracy. To address this, advanced deep learning-based LDCT denoising algorithms have been developed, primarily using Convolutional Neural Networks (CNNs) or Transformer Networks with the Unet architecture. This architecture enhances image detail by integrating feature maps from the encoder and decoder via skip connections. However, current methods often overlook enhancements to the Unet architecture itself, focusing instead on optimizing encoder and decoder structures. This approach can be problematic due to the significant differences in feature map characteristics between the encoder and decoder, where simple fusion strategies may not effectively reconstruct images.In this paper, we introduce WiTUnet, a novel LDCT image denoising method that utilizes nested, dense skip pathways instead of traditional skip connections to improve feature integration. WiTUnet also incorporates a windowed Transformer structure to process images in smaller, non-overlapping segments, reducing computational load. Additionally, the integration of a Local Image Perception Enhancement (LiPe) module in both the encoder and decoder replaces the standard multi-layer perceptron (MLP) in Transformers, enhancing local feature capture and representation. Through extensive experimental comparisons, WiTUnet has demonstrated superior performance over existing methods in key metrics such as Peak Signal-to-Noise Ratio (PSNR), Structural Similarity (SSIM), and Root Mean Square Error (RMSE), significantly improving noise removal and image quality.

Shuang Wang, Xuben Wang, Fei Deng et al.

The extraction of geoelectric structural information from airborne transient electromagnetic(ATEM)data primarily involves data processing and inversion. Conventional methods rely on empirical parameter selection, making it difficult to process complex field data with high noise levels. Additionally, inversion computations are time consuming and often suffer from multiple local minima. Existing deep learning-based approaches separate the data processing steps, where independently trained denoising networks struggle to ensure the reliability of subsequent inversions. Moreover, end to end networks lack interpretability. To address these issues, we propose a unified and interpretable deep learning inversion paradigm based on disentangled representation learning. The network explicitly decomposes noisy data into noise and signal factors, completing the entire data processing workflow based on the signal factors while incorporating physical information for guidance. This approach enhances the network's reliability and interpretability. The inversion results on field data demonstrate that our method can directly use noisy data to accurately reconstruct the subsurface electrical structure. Furthermore, it effectively processes data severely affected by environmental noise, which traditional methods struggle with, yielding improved lateral structural resolution.

Qingsong Yan, Qiang Wang, Kaiyong Zhao et al.

Neural Radiance Fields (NeRF) and 3D Gaussian Splatting (3DGS) have emerged as powerful tools for 3D reconstruction and SLAM tasks. However, their performance depends heavily on accurate camera pose priors. Existing approaches attempt to address this issue by introducing external constraints but fall short of achieving satisfactory accuracy, particularly when camera trajectories are complex. In this paper, we propose a novel method, RA-NeRF, capable of predicting highly accurate camera poses even with complex camera trajectories. Following the incremental pipeline, RA-NeRF reconstructs the scene using NeRF with photometric consistency and incorporates flow-driven pose regulation to enhance robustness during initialization and localization. Additionally, RA-NeRF employs an implicit pose filter to capture the camera movement pattern and eliminate the noise for pose estimation. To validate our method, we conduct extensive experiments on the Tanks\&Temple dataset for standard evaluation, as well as the NeRFBuster dataset, which presents challenging camera pose trajectories. On both datasets, RA-NeRF achieves state-of-the-art results in both camera pose estimation and visual quality, demonstrating its effectiveness and robustness in scene reconstruction under complex pose trajectories.

Shuang Wang, Ming Guo, Xuben Wang et al.

The semi-airborne transient electromagnetic method (SATEM) is capable of conducting rapid surveys over large-scale and hard-to-reach areas. However, the acquired signals are often contaminated by complex noise, which can compromise the accuracy of subsequent inversion interpretations. Traditional denoising techniques primarily rely on parameter selection strategies, which are insufficient for processing field data in noisy environments. With the advent of deep learning, various neural networks have been employed for SATEM signal denoising. However, existing deep learning methods typically use single-mapping learning approaches that struggle to effectively separate signal from noise. These methods capture only partial information and lack interpretability. To overcome these limitations, we propose an interpretable decoupled representation learning framework, termed DREMnet, that disentangles data into content and context factors, enabling robust and interpretable denoising in complex conditions. To address the limitations of CNN and Transformer architectures, we utilize the RWKV architecture for data processing and introduce the Contextual-WKV mechanism, which allows unidirectional WKV to perform bidirectional signal modeling. Our proposed Covering Embedding technique retains the strong local perception of convolutional networks through stacked embedding. Experimental results on test datasets demonstrate that the DREMnet method outperforms existing techniques, with processed field data that more accurately reflects the theoretical signal, offering improved identification of subsurface electrical structures.

Qingsong Yan, Qiang Wang, Kaiyong Zhao et al.

Due to the rapid development of panorama cameras, the task of estimating panorama depth has attracted significant attention from the computer vision community, especially in applications such as robot sensing and autonomous driving. However, existing methods relying on different projection formats often encounter challenges, either struggling with distortion and discontinuity in the case of equirectangular, cubemap, and tangent projections, or experiencing a loss of texture details with the spherical projection. To tackle these concerns, we present SphereFusion, an end-to-end framework that combines the strengths of various projection methods. Specifically, SphereFusion initially employs 2D image convolution and mesh operations to extract two distinct types of features from the panorama image in both equirectangular and spherical projection domains. These features are then projected onto the spherical domain, where a gate fusion module selects the most reliable features for fusion. Finally, SphereFusion estimates panorama depth within the spherical domain. Meanwhile, SphereFusion employs a cache strategy to improve the efficiency of mesh operation. Extensive experiments on three public panorama datasets demonstrate that SphereFusion achieves competitive results with other state-of-the-art methods, while presenting the fastest inference speed at only 17 ms on a 512$\times$1024 panorama image.

Shuang Wang, Xuben Wang, Fei Deng et al.

With the remarkable success of deep learning, applying such techniques to EM methods has emerged as a promising research direction to overcome the limitations of conventional approaches. The effectiveness of deep learning methods depends heavily on the quality of datasets, which directly influences model performance and generalization ability. Existing application studies often construct datasets from random one-dimensional or structurally simple three-dimensional models, which fail to represent the complexity of real geological environments. Furthermore, the absence of standardized, publicly available three-dimensional geoelectric datasets continues to hinder progress in deep learning based EM exploration. To address these limitations, we present OpenEM, a large scale, multi structural three dimensional geoelectric dataset that encompasses a broad range of geologically plausible subsurface structures. OpenEM consists of nine categories of geoelectric models, spanning from simple configurations with anomalous bodies in half space to more complex structures such as flat layers, folded layers, flat faults, curved faults, and their corresponding variants with anomalous bodies. Since three-dimensional forward modeling in electromagnetics is extremely time-consuming, we further developed a deep learning based fast forward modeling approach for OpenEM, enabling efficient and reliable forward modeling across the entire dataset. This capability allows OpenEM to be rapidly deployed for a wide range of tasks. OpenEM provides a unified, comprehensive, and large-scale dataset for common EM exploration systems to accelerate the application of deep learning in electromagnetic methods. The complete dataset, along with the forward modeling codes and trained models, is publicly available at https://doi.org/10.5281/zenodo.17141981.

Peng Fan, Wenping Wang, Fei Deng

The mismatch of speech length and text length poses a challenge in automatic speech recognition (ASR). In previous research, various approaches have been employed to align text with speech, including the utilization of Connectionist Temporal Classification (CTC). In earlier work, a key frame mechanism (KFDS) was introduced, utilizing intermediate CTC outputs to guide downsampling and preserve keyframes, but traditional methods (CTC) failed to align speech and text appropriately when downsampling speech to a text-similar length. In this paper, we focus on speech recognition in those cases where the length of speech aligns closely with that of the corresponding text. To address this issue, we introduce two methods for alignment: a) Time Independence Loss (TIL) and b) Aligned Cross Entropy (AXE) Loss, which is based on edit distance. To enhance the information on keyframes, we incorporate frame fusion by applying weights and summing the keyframe with its context 2 frames. Experimental results on AISHELL-1 and AISHELL-2 dataset subsets show that the proposed methods outperform the previous work and achieve a reduction of at least 86\% in the number of frames.

Bin Wang, Fei Deng, Zeyu Chen et al.

Source-Free Domain Adaptation (SFDA) enables domain adaptation for semantic segmentation of Remote Sensing Images (RSIs) using only a well-trained source model and unlabeled target domain data. However, the lack of ground-truth labels in the target domain often leads to the generation of noisy pseudo-labels. Such noise impedes the effective mitigation of domain shift (DS). To address this challenge, we propose ProSFDA, a prototype-guided SFDA framework. It employs prototype-weighted pseudo-labels to facilitate reliable self-training (ST) under pseudo-labels noise. We, in addition, introduce a prototype-contrast strategy that encourages the aggregation of features belonging to the same class, enabling the model to learn discriminative target domain representations without relying on ground-truth supervision. Extensive experiments show that our approach substantially outperforms existing methods.

Teng Xiao, Qi Hu, Qingsong Yan et al.

Pan-Tilt-Zoom (PTZ) cameras with wide-angle lenses are widely used in surveillance but often require image rectification due to their inherent nonlinear distortions. Current deep learning approaches typically struggle to maintain fine-grained geometric details, resulting in inaccurate rectification. This paper presents a Forward Distortion and Backward Warping Network (FDBW-Net), a novel framework for wide-angle image rectification. It begins by using a forward distortion model to synthesize barrel-distorted images, reducing pixel redundancy and preventing blur. The network employs a pyramid context encoder with attention mechanisms to generate backward warping flows containing geometric details. Then, a multi-scale decoder is used to restore distorted features and output rectified images. FDBW-Net's performance is validated on diverse datasets: public benchmarks, AirSim-rendered PTZ camera imagery, and real-scene PTZ camera datasets. It demonstrates that FDBW-Net achieves SOTA performance in distortion rectification, boosting the adaptability of PTZ cameras for practical visual applications.

Shuang Wang, Fei Deng, Peifan Jiang et al.

Due to limitations such as geographic, physical, or economic factors, collected seismic data often have missing traces. Traditional seismic data reconstruction methods face the challenge of selecting numerous empirical parameters and struggle to handle large-scale continuous missing traces. With the advancement of deep learning, various diffusion models have demonstrated strong reconstruction capabilities. However, these UNet-based diffusion models require significant computational resources and struggle to learn the correlation between different traces in seismic data. To address the complex and irregular missing situations in seismic data, we propose a latent diffusion transformer utilizing representation learning for seismic data reconstruction. By employing a mask modeling scheme based on representation learning, the representation module uses the token sequence of known data to infer the token sequence of unknown data, enabling the reconstructed data from the diffusion model to have a more consistent data distribution and better correlation and accuracy with the known data. We propose the Representation Diffusion Transformer architecture, and a relative positional bias is added when calculating attention, enabling the diffusion model to achieve global modeling capability for seismic data. Using a pre-trained data compression model compresses the training and inference processes of the diffusion model into a latent space, which, compared to other diffusion model-based reconstruction methods, reduces computational and inference costs. Reconstruction experiments on field and synthetic datasets indicate that our method achieves higher reconstruction accuracy than existing methods and can handle various complex missing scenarios.

Peifan Jiang, Xuben Wang, Shuang Wang et al.

Magnetotelluric deep learning (DL) inversion methods based on joint data-driven and physics-driven have become a hot topic in recent years. When mapping observation data (or forward modeling data) to the resistivity model using neural networks (NNs), incorporating the error (loss) term of the inversion resistivity's forward modeling response--which introduces physical information about electromagnetic field propagation--can significantly enhance the inversion accuracy. To efficiently achieve data-physical dual-driven MT deep learning inversion for large-scale 3-D MT data, we propose using DL forward modeling networks to compute this portion of the loss. This approach introduces pseudo-physical information through the forward modeling of NN simulation, further guiding the inversion network fitting. Specifically, we first pre-train the forward modeling networks as fixed forward modeling operators, then transfer and integrate them into the inversion network training, and finally optimize the inversion network by minimizing the multinomial loss. Theoretical experimental results indicate that despite some simulation errors in DL forward modeling, the introduced pseudo-physical information still enhances inversion accuracy and significantly mitigates the overfitting problem during training. Additionally, we propose a new input mode that involves masking and adding noise to the data, simulating the field data environment of 3-D MT inversion, thereby making the method more flexible and effective for practical applications.

Jindong Jiang, Fei Deng, Gautam Singh et al.

Recent State Space Models (SSMs) such as S4, S5, and Mamba have shown remarkable computational benefits in long-range temporal dependency modeling. However, in many sequence modeling problems, the underlying process is inherently modular and it is of interest to have inductive biases that mimic this modular structure. In this paper, we introduce SlotSSMs, a novel framework for incorporating independent mechanisms into SSMs to preserve or encourage separation of information. Unlike conventional SSMs that maintain a monolithic state vector, SlotSSMs maintains the state as a collection of multiple vectors called slots. Crucially, the state transitions are performed independently per slot with sparse interactions across slots implemented via the bottleneck of self-attention. In experiments, we evaluate our model in object-centric learning, 3D visual reasoning, and long-context video understanding tasks, which involve modeling multiple objects and their long-range temporal dependencies. We find that our proposed design offers substantial performance gains over existing sequence modeling methods. Project page is available at https://slotssms.github.io/

Chang Chen, Junyeob Baek, Fei Deng et al.

Despite the recent advancements in offline RL, no unified algorithm could achieve superior performance across a broad range of tasks. Offline \textit{value function learning}, in particular, struggles with sparse-reward, long-horizon tasks due to the difficulty of solving credit assignment and extrapolation errors that accumulates as the horizon of the task grows.~On the other hand, models that can perform well in long-horizon tasks are designed specifically for goal-conditioned tasks, which commonly perform worse than value function learning methods on short-horizon, dense-reward scenarios. To bridge this gap, we propose a hierarchical planner designed for offline RL called PlanDQ. PlanDQ incorporates a diffusion-based planner at the high level, named D-Conductor, which guides the low-level policy through sub-goals. At the low level, we used a Q-learning based approach called the Q-Performer to accomplish these sub-goals. Our experimental results suggest that PlanDQ can achieve superior or competitive performance on D4RL continuous control benchmark tasks as well as AntMaze, Kitchen, and Calvin as long-horizon tasks.

Gautam Singh, Fei Deng, Sungjin Ahn

Although DALL-E has shown an impressive ability of composition-based systematic generalization in image generation, it requires the dataset of text-image pairs and the compositionality is provided by the text. In contrast, object-centric representation models like the Slot Attention model learn composable representations without the text prompt. However, unlike DALL-E its ability to systematically generalize for zero-shot generation is significantly limited. In this paper, we propose a simple but novel slot-based autoencoding architecture, called SLATE, for combining the best of both worlds: learning object-centric representations that allows systematic generalization in zero-shot image generation without text. As such, this model can also be seen as an illiterate DALL-E model. Unlike the pixel-mixture decoders of existing object-centric representation models, we propose to use the Image GPT decoder conditioned on the slots for capturing complex interactions among the slots and pixels. In experiments, we show that this simple and easy-to-implement architecture not requiring a text prompt achieves significant improvement in in-distribution and out-of-distribution (zero-shot) image generation and qualitatively comparable or better slot-attention structure than the models based on mixture decoders.

Chang Chen, Fei Deng, Sungjin Ahn

A crucial ability of human intelligence is to build up models of individual 3D objects from partial scene observations. Recent works achieve object-centric generation but without the ability to infer the representation, or achieve 3D scene representation learning but without object-centric compositionality. Therefore, learning to represent and render 3D scenes with object-centric compositionality remains elusive. In this paper, we propose a probabilistic generative model for learning to build modular and compositional 3D object models from partial observations of a multi-object scene. The proposed model can (i) infer the 3D object representations by learning to search and group object areas and also (ii) render from an arbitrary viewpoint not only individual objects but also the full scene by compositing the objects. The entire learning process is unsupervised and end-to-end. In experiments, in addition to generation quality, we also demonstrate that the learned representation permits object-wise manipulation and novel scene generation, and generalizes to various settings. Results can be found on our project website: https://sites.google.com/view/roots3d

Zhixuan Lin, Yi-Fu Wu, Skand Vishwanath Peri et al.

The ability to decompose complex multi-object scenes into meaningful abstractions like objects is fundamental to achieve higher-level cognition. Previous approaches for unsupervised object-oriented scene representation learning are either based on spatial-attention or scene-mixture approaches and limited in scalability which is a main obstacle towards modeling real-world scenes. In this paper, we propose a generative latent variable model, called SPACE, that provides a unified probabilistic modeling framework that combines the best of spatial-attention and scene-mixture approaches. SPACE can explicitly provide factorized object representations for foreground objects while also decomposing background segments of complex morphology. Previous models are good at either of these, but not both. SPACE also resolves the scalability problems of previous methods by incorporating parallel spatial-attention and thus is applicable to scenes with a large number of objects without performance degradations. We show through experiments on Atari and 3D-Rooms that SPACE achieves the above properties consistently in comparison to SPAIR, IODINE, and GENESIS. Results of our experiments can be found on our project website: https://sites.google.com/view/space-project-page

Qiang Wang, Shizhen Zheng, Qingsong Yan et al.

Indoor robotics localization, navigation, and interaction heavily rely on scene understanding and reconstruction. Compared to the monocular vision which usually does not explicitly introduce any geometrical constraint, stereo vision-based schemes are more promising and robust to produce accurate geometrical information, such as surface normal and depth/disparity. Besides, deep learning models trained with large-scale datasets have shown their superior performance in many stereo vision tasks. However, existing stereo datasets rarely contain the high-quality surface normal and disparity ground truth, which hardly satisfies the demand of training a prospective deep model for indoor scenes. To this end, we introduce a large-scale synthetic but naturalistic indoor robotics stereo (IRS) dataset with over 100K stereo RGB images and high-quality surface normal and disparity maps. Leveraging the advanced rendering techniques of our customized rendering engine, the dataset is considerably close to the real-world captured images and covers several visual effects, such as brightness changes, light reflection/transmission, lens flare, vivid shadow, etc. We compare the data distribution of IRS with existing stereo datasets to illustrate the typical visual attributes of indoor scenes. Besides, we present DTN-Net, a two-stage deep model for surface normal estimation. Extensive experiments show the advantages and effectiveness of IRS in training deep models for disparity estimation, and DTN-Net provides state-of-the-art results for normal estimation compared to existing methods.

Fei Deng, Zhuo Zhi, Sungjin Ahn

Compositional structures between parts and objects are inherent in natural scenes. Modeling such compositional hierarchies via unsupervised learning can bring various benefits such as interpretability and transferability, which are important in many downstream tasks. In this paper, we propose the first deep latent variable model, called RICH, for learning Representation of Interpretable Compositional Hierarchies. At the core of RICH is a latent scene graph representation that organizes the entities of a scene into a tree structure according to their compositional relationships. During inference, taking top-down approach, RICH is able to use higher-level representation to guide lower-level decomposition. This avoids the difficult problem of routing between parts and objects that is faced by bottom-up approaches. In experiments on images containing multiple objects with different part compositions, we demonstrate that RICH is able to learn the latent compositional hierarchy and generate imaginary scenes.

Fei Deng, Jinsheng Ren, Feng Chen

There has been a gap between artificial intelligence and human intelligence. In this paper, we identify three key elements forming human intelligence, and suggest that abstraction learning combines these elements and is thus a way to bridge the gap. Prior researches in artificial intelligence either specify abstraction by human experts, or take abstraction as a qualitative explanation for the model. This paper aims to learn abstraction directly. We tackle three main challenges: representation, objective function, and learning algorithm. Specifically, we propose a partition structure that contains pre-allocated abstraction neurons; we formulate abstraction learning as a constrained optimization problem, which integrates abstraction properties; we develop a network evolution algorithm to solve this problem. This complete framework is named ONE (Optimization via Network Evolution). In our experiments on MNIST, ONE shows elementary human-like intelligence, including low energy consumption, knowledge sharing, and lifelong learning.