Yuting Gao

Yuting Gao, Jinfeng Liu, Zihan Xu et al. · tencent-ai

Large-scale vision-language pre-training has achieved promising results on downstream tasks. Existing methods highly rely on the assumption that the image-text pairs crawled from the Internet are in perfect one-to-one correspondence. However, in real scenarios, this assumption can be difficult to hold: the text description, obtained by crawling the affiliated metadata of the image, often suffers from the semantic mismatch and the mutual compatibility. To address these issues, we introduce PyramidCLIP, which constructs an input pyramid with different semantic levels for each modality, and aligns visual elements and linguistic elements in the form of hierarchy via peer-level semantics alignment and cross-level relation alignment. Furthermore, we soften the loss of negative samples (unpaired samples) so as to weaken the strict constraint during the pre-training stage, thus mitigating the risk of forcing the model to distinguish compatible negative pairs. Experiments on five downstream tasks demonstrate the effectiveness of the proposed PyramidCLIP. In particular, with the same amount of 15 million pre-training image-text pairs, PyramidCLIP exceeds CLIP on ImageNet zero-shot classification top-1 accuracy by 10.6%/13.2%/10.0% with ResNet50/ViT-B32/ViT-B16 based image encoder respectively. When scaling to larger datasets, PyramidCLIP achieves the state-of-the-art results on several downstream tasks. In particular, the results of PyramidCLIP-ResNet50 trained on 143M image-text pairs surpass that of CLIP using 400M data on ImageNet zero-shot classification task, significantly improving the data efficiency of CLIP.

Peixian Chen, Mengdan Zhang, Yunhang Shen et al. · tencent-ai

Vision transformers (ViTs) are changing the landscape of object detection approaches. A natural usage of ViTs in detection is to replace the CNN-based backbone with a transformer-based backbone, which is straightforward and effective, with the price of bringing considerable computation burden for inference. More subtle usage is the DETR family, which eliminates the need for many hand-designed components in object detection but introduces a decoder demanding an extra-long time to converge. As a result, transformer-based object detection can not prevail in large-scale applications. To overcome these issues, we propose a novel decoder-free fully transformer-based (DFFT) object detector, achieving high efficiency in both training and inference stages, for the first time. We simplify objection detection into an encoder-only single-level anchor-based dense prediction problem by centering around two entry points: 1) Eliminate the training-inefficient decoder and leverage two strong encoders to preserve the accuracy of single-level feature map prediction; 2) Explore low-level semantic features for the detection task with limited computational resources. In particular, we design a novel lightweight detection-oriented transformer backbone that efficiently captures low-level features with rich semantics based on a well-conceived ablation study. Extensive experiments on the MS COCO benchmark demonstrate that DFFT_SMALL outperforms DETR by 2.5% AP with 28% computation cost reduction and more than $10$x fewer training epochs. Compared with the cutting-edge anchor-based detector RetinaNet, DFFT_SMALL obtains over 5.5% AP gain while cutting down 70% computation cost.

Yuting Gao, Jinfeng Liu, Zihan Xu et al.

During the preceding biennium, vision-language pre-training has achieved noteworthy success on several downstream tasks. Nevertheless, acquiring high-quality image-text pairs, where the pairs are entirely exclusive of each other, remains a challenging task, and noise exists in the commonly used datasets. To address this issue, we propose SoftCLIP, a novel approach that relaxes the strict one-to-one constraint and achieves a soft cross-modal alignment by introducing a softened target, which is generated from the fine-grained intra-modal self-similarity. The intra-modal guidance is indicative to enable two pairs have some local similarities and model many-to-many relationships between the two modalities. Besides, since the positive still dominates in the softened target distribution, we disentangle the negatives in the distribution to further boost the relation alignment with the negatives in the cross-modal learning. Extensive experiments demonstrate the effectiveness of SoftCLIP. In particular, on ImageNet zero-shot classification task, using CC3M/CC12M as pre-training dataset, SoftCLIP brings a top-1 accuracy improvement of 6.8%/7.2% over the CLIP baseline.

Tong Wu, Yulei Qin, Enwei Zhang et al.

Retrieval augmentation has become an effective solution to empower large language models (LLMs) with external and verified knowledge sources from the database, which overcomes the limitations and hallucinations of LLMs in handling up-to-date and domain-specific information. However, existing embedding models for text retrieval usually have three non-negligible limitations. First, the number and diversity of samples in a batch are too restricted to supervise the modeling of textual nuances at scale. Second, the high proportional noise are detrimental to the semantic correctness and consistency of embeddings. Third, the equal treatment to easy and difficult samples would cause sub-optimum convergence of embeddings with poorer generalization. In this paper, we propose the PEG, a progressively learned embeddings for robust text retrieval. Specifically, we increase the training in-batch negative samples to 80,000, and for each query, we extracted five hard negatives. Concurrently, we incorporated a progressive learning mechanism, enabling the model to dynamically modulate its attention to the samples throughout the entire training process. Additionally, PEG is trained on more than 100 million data, encompassing a wide range of domains (e.g., finance, medicine, and tourism) and covering various tasks (e.g., question-answering, machine reading comprehension, and similarity matching). Extensive experiments conducted on C-MTEB and DuReader demonstrate that PEG surpasses state-of-the-art embeddings in retrieving true positives, highlighting its significant potential for applications in LLMs. Our model is publicly available at https://huggingface.co/TownsWu/PEG.

Inclusion AI, Biao Gong, Cheng Zou et al.

We propose Ming-Omni, a unified multimodal model capable of processing images, text, audio, and video, while demonstrating strong proficiency in both speech and image generation. Ming-Omni employs dedicated encoders to extract tokens from different modalities, which are then processed by Ling, an MoE architecture equipped with newly proposed modality-specific routers. This design enables a single model to efficiently process and fuse multimodal inputs within a unified framework, thereby facilitating diverse tasks without requiring separate models, task-specific fine-tuning, or structural redesign. Importantly, Ming-Omni extends beyond conventional multimodal models by supporting audio and image generation. This is achieved through the integration of an advanced audio decoder for natural-sounding speech and Ming-Lite-Uni for high-quality image generation, which also allow the model to engage in context-aware chatting, perform text-to-speech conversion, and conduct versatile image editing. Our experimental results showcase Ming-Omni offers a powerful solution for unified perception and generation across all modalities. Notably, our proposed Ming-Omni is the first open-source model we are aware of to match GPT-4o in modality support, and we release all code and model weights to encourage further research and development in the community.

Tao Chen, Enwei Zhang, Yuting Gao et al.

Although In-Context Learning (ICL) brings remarkable performance gains to Large Language Models (LLMs), the improvements remain lower than fine-tuning on downstream tasks. This paper introduces Multi-Modal In-Context Tuning (MMICT), a novel multi-modal fine-tuning paradigm that boosts multi-modal fine-tuning by fully leveraging the promising ICL capability of multi-modal LLMs (MM-LLMs). We propose the Multi-Modal Hub (M-Hub), a unified module that captures various multi-modal features according to different inputs and objectives. Based on M-Hub, MMICT enables MM-LLMs to learn from in-context visual-guided textual features and subsequently generate outputs conditioned on the textual-guided visual features. Moreover, leveraging the flexibility of M-Hub, we design a variety of in-context demonstrations. Extensive experiments on a diverse range of downstream multi-modal tasks demonstrate that MMICT significantly outperforms traditional fine-tuning strategy and the vanilla ICT method that directly takes the concatenation of all information from different modalities as input. Our implementation is available at: https://github.com/KDEGroup/MMICT.

Xiaoyang Chen, Xinan Dai, Yu Du et al.

To advance the mathematical proficiency of large language models (LLMs), the DeepMath team has launched an open-source initiative aimed at developing an open mathematical LLM and systematically evaluating its mathematical creativity. This paper represents the initial contribution of this initiative. While recent developments in mathematical LLMs have predominantly emphasized reasoning skills, as evidenced by benchmarks on elementary to undergraduate-level mathematical tasks, the creative capabilities of these models have received comparatively little attention, and evaluation datasets remain scarce. To address this gap, we propose an evaluation criteria for mathematical creativity and introduce DeepMath-Creative, a novel, high-quality benchmark comprising constructive problems across algebra, geometry, analysis, and other domains. We conduct a systematic evaluation of mainstream LLMs' creative problem-solving abilities using this dataset. Experimental results show that even under lenient scoring criteria -- emphasizing core solution components and disregarding minor inaccuracies, such as small logical gaps, incomplete justifications, or redundant explanations -- the best-performing model, O3 Mini, achieves merely 70% accuracy, primarily on basic undergraduate-level constructive tasks. Performance declines sharply on more complex problems, with models failing to provide substantive strategies for open problems. These findings suggest that, although current LLMs display a degree of constructive proficiency on familiar and lower-difficulty problems, such performance is likely attributable to the recombination of memorized patterns rather than authentic creative insight or novel synthesis.

Yuting Gao, Jia-Xin Zhuang, Shaohui Lin et al.

While self-supervised representation learning (SSL) has received widespread attention from the community, recent research argue that its performance will suffer a cliff fall when the model size decreases. The current method mainly relies on contrastive learning to train the network and in this work, we propose a simple yet effective Distilled Contrastive Learning (DisCo) to ease the issue by a large margin. Specifically, we find the final embedding obtained by the mainstream SSL methods contains the most fruitful information, and propose to distill the final embedding to maximally transmit a teacher's knowledge to a lightweight model by constraining the last embedding of the student to be consistent with that of the teacher. In addition, in the experiment, we find that there exists a phenomenon termed Distilling BottleNeck and present to enlarge the embedding dimension to alleviate this problem. Our method does not introduce any extra parameter to lightweight models during deployment. Experimental results demonstrate that our method achieves the state-of-the-art on all lightweight models. Particularly, when ResNet-101/ResNet-50 is used as teacher to teach EfficientNet-B0, the linear result of EfficientNet-B0 on ImageNet is very close to ResNet-101/ResNet-50, but the number of parameters of EfficientNet-B0 is only 9.4\%/16.3\% of ResNet-101/ResNet-50. Code is available at https://github. com/Yuting-Gao/DisCo-pytorch.

Hao Cheng, Fanxu Meng, Ke Li et al.

Filter is the key component in modern convolutional neural networks (CNNs). However, since CNNs are usually over-parameterized, a pre-trained network always contain some invalid (unimportant) filters. These filters have relatively small $l_{1}$ norm and contribute little to the output (\textbf{Reason}). While filter pruning removes these invalid filters for efficiency consideration, we tend to reactivate them to improve the representation capability of CNNs. In this paper, we introduce filter grafting (\textbf{Method}) to achieve this goal. The activation is processed by grafting external information (weights) into invalid filters. To better perform the grafting, we develop a novel criterion to measure the information of filters and an adaptive weighting strategy to balance the grafted information among networks. After the grafting operation, the network has fewer invalid filters compared with its initial state, enpowering the model with more representation capacity. Meanwhile, since grafting is operated reciprocally on all networks involved, we find that grafting may lose the information of valid filters when improving invalid filters. To gain a universal improvement on both valid and invalid filters, we compensate grafting with distillation (\textbf{Cultivation}) to overcome the drawback of grafting . Extensive experiments are performed on the classification and recognition tasks to show the superiority of our method. Code is available at \textcolor{black}{\emph{https://github.com/fxmeng/filter-grafting}}.

Xiao Cui, Yulei Qin, Yuting Gao et al.

Knowledge distillation (KD) has been widely adopted to compress large language models (LLMs). Existing KD methods investigate various divergence measures including the Kullback-Leibler (KL), reverse Kullback-Leibler (RKL), and Jensen-Shannon (JS) divergences. However, due to limitations inherent in their assumptions and definitions, these measures fail to deliver effective supervision when few distribution overlap exists between the teacher and the student. In this paper, we show that the aforementioned KL, RKL, and JS divergences respectively suffer from issues of mode-averaging, mode-collapsing, and mode-underestimation, which deteriorates logits-based KD for diverse NLP tasks. We propose the Sinkhorn Knowledge Distillation (SinKD) that exploits the Sinkhorn distance to ensure a nuanced and precise assessment of the disparity between teacher and student distributions. Besides, profit by properties of the Sinkhorn metric, we can get rid of sample-wise KD that restricts the perception of divergence in each teacher-student sample pair. Instead, we propose a batch-wise reformulation to capture geometric intricacies of distributions across samples in the high-dimensional space. Comprehensive evaluation on GLUE and SuperGLUE, in terms of comparability, validity, and generalizability, highlights our superiority over state-of-the-art methods on all kinds of LLMs with encoder-only, encoder-decoder, and decoder-only architectures.

Wensheng Pan, Timin Gao, Yan Zhang et al.

Image Quality Assessment (IQA) models benefit significantly from semantic information, which allows them to treat different types of objects distinctly. Currently, leveraging semantic information to enhance IQA is a crucial research direction. Traditional methods, hindered by a lack of sufficiently annotated data, have employed the CLIP image-text pretraining model as their backbone to gain semantic awareness. However, the generalist nature of these pre-trained Vision-Language (VL) models often renders them suboptimal for IQA-specific tasks. Recent approaches have attempted to address this mismatch using prompt technology, but these solutions have shortcomings. Existing prompt-based VL models overly focus on incremental semantic information from text, neglecting the rich insights available from visual data analysis. This imbalance limits their performance improvements in IQA tasks. This paper introduces an innovative multi-modal prompt-based methodology for IQA. Our approach employs carefully crafted prompts that synergistically mine incremental semantic information from both visual and linguistic data. Specifically, in the visual branch, we introduce a multi-layer prompt structure to enhance the VL model's adaptability. In the text branch, we deploy a dual-prompt scheme that steers the model to recognize and differentiate between scene category and distortion type, thereby refining the model's capacity to assess image quality. Our experimental findings underscore the effectiveness of our method over existing Blind Image Quality Assessment (BIQA) approaches. Notably, it demonstrates competitive performance across various datasets. Our method achieves Spearman Rank Correlation Coefficient (SRCC) values of 0.961(surpassing 0.946 in CSIQ) and 0.941 (exceeding 0.930 in KADID), illustrating its robustness and accuracy in diverse contexts.

Inclusion AI, Bowen Ma, Cheng Zou et al.

We propose Ming-Flash-Omni, an upgraded version of Ming-Omni, built upon a sparser Mixture-of-Experts (MoE) variant of Ling-Flash-2.0 with 100 billion total parameters, of which only 6.1 billion are active per token. This architecture enables highly efficient scaling (dramatically improving computational efficiency while significantly expanding model capacity) and empowers stronger unified multimodal intelligence across vision, speech, and language, representing a key step toward Artificial General Intelligence (AGI). Compared to its predecessor, the upgraded version exhibits substantial improvements across multimodal understanding and generation. We significantly advance speech recognition capabilities, achieving state-of-the-art performance in contextual ASR and highly competitive results in dialect-aware ASR. In image generation, Ming-Flash-Omni introduces high-fidelity text rendering and demonstrates marked gains in scene consistency and identity preservation during image editing. Furthermore, Ming-Flash-Omni introduces generative segmentation, a capability that not only achieves strong standalone segmentation performance but also enhances spatial control in image generation and improves editing consistency. Notably, Ming-Flash-Omni achieves state-of-the-art results in text-to-image generation and generative segmentation, and sets new records on all 12 contextual ASR benchmarks, all within a single unified architecture.

Linglin Jing, Yuting Gao, Zhigang Wang et al.

Recent advancements have shown that the Mixture of Experts (MoE) approach significantly enhances the capacity of large language models (LLMs) and improves performance on downstream tasks. Building on these promising results, multi-modal large language models (MLLMs) have increasingly adopted MoE techniques. However, existing multi-modal MoE tuning methods typically face two key challenges: expert uniformity and router rigidity. Expert uniformity occurs because MoE experts are often initialized by simply replicating the FFN parameters from LLMs, leading to homogenized expert functions and weakening the intended diversification of the MoE architecture. Meanwhile, router rigidity stems from the prevalent use of static linear routers for expert selection, which fail to distinguish between visual and textual tokens, resulting in similar expert distributions for image and text. To address these limitations, we propose EvoMoE, an innovative MoE tuning framework. EvoMoE introduces a meticulously designed expert initialization strategy that progressively evolves multiple robust experts from a single trainable expert, a process termed expert evolution that specifically targets severe expert homogenization. Furthermore, we introduce the Dynamic Token-aware Router (DTR), a novel routing mechanism that allocates input tokens to appropriate experts based on their modality and intrinsic token values. This dynamic routing is facilitated by hypernetworks, which dynamically generate routing weights tailored for each individual token. Extensive experiments demonstrate that EvoMoE significantly outperforms other sparse MLLMs across a variety of multi-modal benchmarks, including MME, MMBench, TextVQA, and POPE. Our results highlight the effectiveness of EvoMoE in enhancing the performance of MLLMs by addressing the critical issues of expert uniformity and router rigidity.

Ruihan Xu, Yuting Gao, Lan Wang et al.

Large Multimodal Models (LMMs) have achieved remarkable success in vision-language tasks, yet their vast parameter counts are often underutilized during both training and inference. In this work, we embrace the idea of looping back to move forward: reusing model parameters through recursive refinement to extract stronger multimodal representations without increasing model size. We propose RecursiveVLM, a recursive Transformer architecture tailored for LMMs. Two key innovations enable effective looping: (i) a Recursive Connector that aligns features across recursion steps by fusing intermediate-layer hidden states and applying modality-specific projections, respecting the distinct statistical structures of vision and language tokens; (ii) a Monotonic Recursion Loss that supervises every step and guarantees performance improves monotonically with recursion depth. This design transforms recursion into an on-demand refinement mechanism: delivering strong results with few loops on resource-constrained devices and progressively improving outputs when more computation resources are available. Experiments show consistent gains of +3% over standard Transformers and +7% over vanilla recursive baselines, demonstrating that strategic looping is a powerful path toward efficient, deployment-adaptive LMMs.

Yuting Gao, Weihao Chen, Lan Wang et al.

Preference learning has recently emerged as a pivotal strategy for post-training alignment of Multimodal Large Language Models (MLLMs). However, existing approaches predominantly rely on external human-annotated preference data, which is costly and labor-intensive to collect. In this work, we propose OrdMoE, a novel preference alignment framework that bypasses the reliance on external human preferences entirely by leveraging intrinsic signals within Mixture-of-Experts (MoE) architectures. Specifically, we observe that the router's expert selection scores implicitly encode a quality-aware ranking of responses (i.e. higher-scoring experts consistently generate higher-quality outputs). Building on this insight, OrdMoE constructs an internal preference hierarchy by grouping experts into ranked tiers based on their per-token routing scores and activating each tier separately to produce a sequence of responses with increasing quality. This yields a zero-cost, self-supervised preference ordering over generated responses, which can be directly optimized using standard preference learning objectives. Extensive experiments across multiple multimodal benchmarks demnstrate that OrdMoE significantly enhances both alignment and overall performance of multimodal Mixture-of-Experts LLMs, achieving competitive results without requiring any human-annotated preference data.

Yuting Gao, Wang Lan, Hengyuan Zhao et al.

Multimodal Mixture-of-Experts (MoE) models offer a promising path toward scalable and efficient large vision-language systems. However, existing approaches rely on rigid routing strategies (typically activating a fixed number of experts per token) ignoring the inherent heterogeneity in semantic importance across modalities. This leads to suboptimal compute allocation, where redundant tokens consume as many resources as critical ones. To address this, we propose AnyExperts, a novel on-demand, budget-aware dynamic routing framework that allocates a variable total number of expert slots per token based on its semantic importance. Crucially, to prevent uncontrolled compute growth, the total slots per token are constrained within a fixed range, and each slot is filled by either a real expert or a virtual expert, with the virtual share capped at a small maximum (e.g., 20%). The model then adaptively balances the real-to-virtual ratio per token, assigning more real experts to semantically rich regions and relying more on virtual experts for redundant content. Evaluated across diverse tasks in visual understanding, audio understanding, and NLP understanding, AnyExperts improves performance under the same compute budget. Notably, on general image/video tasks, it achieves comparable accuracy with 40% fewer real expert activations; on text-dense tasks (OCR and NLP), it maintains performance while reducing real expert usage by 10%. These results demonstrate that fine-grained, importance-driven expert allocation significantly enhances both the efficiency and effectiveness of multimodal MoE models.

Honglin Li, Yuting Gao, Chenglu Zhu et al.

Multimodal large language models (MLLMs) are closing the gap to human visual perception capability rapidly, while, still lag behind on attending to subtle images details or locating small objects precisely, etc. Common schemes to tackle these issues include deploying multiple vision encoders or operating on original high-resolution images. Few studies have concentrated on taking the textual instruction into improving visual representation, resulting in losing focus in some vision-centric tasks, a phenomenon we herein termed as Amblyopia. In this work, we introduce Panther, a MLLM that closely adheres to user instruction and locates targets of interests precisely, with the finesse of a black panther. Specifically, Panther comprises three integral components: Panther-VE, Panther-Bridge, and Panther-Decoder. Panther-VE integrates user instruction information at the early stages of the vision encoder, thereby extracting the most relevant and useful visual representations. The Panther-Bridge module, equipped with powerful filtering capabilities, significantly reduces redundant visual information, leading to a substantial savings in training costs. The Panther-Decoder is versatile and can be employed with any decoder-only architecture of LLMs without discrimination. Experimental results, particularly on vision-centric benchmarks, have demonstrated the effectiveness of Panther.

Yuncheng Yang, Chuyan Zhang, Zuopeng Yang et al.

Prompt learning is effective for fine-tuning foundation models to improve their generalization across a variety of downstream tasks. However, the prompts that are independently optimized along a single modality path, may sacrifice the vision-language alignment of pre-trained models in return for improved performance on specific tasks and classes, leading to poorer generalization. In this paper, we first demonstrate that prompt tuning along only one single branch of CLIP (e.g., language or vision) is the reason why the misalignment occurs. Without proper regularization across the learnable parameters in different modalities, prompt learning violates the original pre-training constraints inherent in the two-tower architecture. To address such misalignment, we first propose feature shift, which is defined as the variation of embeddings after introducing the learned prompts, to serve as an explanatory tool. We dive into its relation with generalizability and thereafter propose RESTORE, a multi-modal prompt learning method that exerts explicit constraints on cross-modal consistency. To be more specific, to prevent feature misalignment, a feature shift consistency is introduced to synchronize inter-modal feature shifts by measuring and regularizing the magnitude of discrepancy during prompt tuning. In addition, we propose a "surgery" block to avoid short-cut hacking, where cross-modal misalignment can still be severe if the feature shift of each modality varies drastically at the same rate. It is implemented as feed-forward adapters upon both modalities to alleviate the misalignment problem. Extensive experiments on 15 datasets demonstrate that our method outperforms the state-of-the-art prompt tuning methods without compromising feature alignment.

Huixiang Luo, Hao Cheng, Fanxu Meng et al.

Pseudo-labeling (PL) and Data Augmentation-based Consistency Training (DACT) are two approaches widely used in Semi-Supervised Learning (SSL) methods. These methods exhibit great power in many machine learning tasks by utilizing unlabeled data for efficient training. But in a more realistic setting (termed as open-set SSL), where unlabeled dataset contains out-of-distribution (OOD) samples, the traditional SSL methods suffer severe performance degradation. Recent approaches mitigate the negative influence of OOD samples by filtering them out from the unlabeled data. However, it is not clear whether directly removing the OOD samples is the best choice. Furthermore, why PL and DACT could perform differently in open-set SSL remains a mystery. In this paper, we thoroughly analyze various SSL methods (PL and DACT) on open-set SSL and discuss pros and cons of these two approaches separately. Based on our analysis, we propose Style Disturbance to improve traditional SSL methods on open-set SSL and experimentally show our approach can achieve state-of-the-art results on various datasets by utilizing OOD samples properly. We believe our study can bring new insights for SSL research.

Yi Gu, Jie Li, Yuting Gao et al.

Neural networks are susceptible to catastrophic forgetting. They fail to preserve previously acquired knowledge when adapting to new tasks. Inspired by human associative memory system, we propose a brain-like approach that imitates the associative learning process to achieve continual learning. We design a heuristics mechanism to potentiatively stimulate the model, which guides the model to recall the historical episodes based on the current circumstance and obtained association experience. Besides, a distillation measure is added to depressively alter the efficacy of synaptic transmission, which dampens the feature reconstruction learning for new task. The framework is mediated by potentiation and depression stimulation that play opposing roles in directing synaptic and behavioral plasticity. It requires no access to the original data and is more similar to human cognitive process. Experiments demonstrate the effectiveness of our method in alleviating catastrophic forgetting on image-to-image translation tasks.

Yi Gu, Yuting Gao, Jie Li et al.

Liquify is a common technique for image editing, which can be used for image distortion. Due to the uncertainty in the distortion variation, restoring distorted images caused by liquify filter is a challenging task. To edit images in an efficient way, distorted images are expected to be restored automatically. This paper aims at the distorted image restoration, which is characterized by seeking the appropriate warping and completion of a distorted image. Existing methods focus on the hardware assistance or the geometric principle to solve the specific regular deformation caused by natural phenomena, but they cannot handle the irregularity and uncertainty of artificial distortion in this task. To address this issue, we propose a novel generative and discriminative learning method based on deep neural networks, which can learn various reconstruction mappings and represent complex and high-dimensional data. This method decomposes the task into a rectification stage and a refinement stage. The first stage generative network predicts the mapping from the distorted images to the rectified ones. The second stage generative network then further optimizes the perceptual quality. Since there is no available dataset or benchmark to explore this task, we create a Distorted Face Dataset (DFD) by forward distortion mapping based on CelebA dataset. Extensive experimental evaluation on the proposed benchmark and the application demonstrates that our method is an effective way for distorted image restoration.

Jinpeng Wang, Yuting Gao, Ke Li et al.

Self-supervised learning has shown great potentials in improving the video representation ability of deep neural networks by getting supervision from the data itself. However, some of the current methods tend to cheat from the background, i.e., the prediction is highly dependent on the video background instead of the motion, making the model vulnerable to background changes. To mitigate the model reliance towards the background, we propose to remove the background impact by adding the background. That is, given a video, we randomly select a static frame and add it to every other frames to construct a distracting video sample. Then we force the model to pull the feature of the distracting video and the feature of the original video closer, so that the model is explicitly restricted to resist the background influence, focusing more on the motion changes. We term our method as \emph{Background Erasing} (BE). It is worth noting that the implementation of our method is so simple and neat and can be added to most of the SOTA methods without much efforts. Specifically, BE brings 16.4% and 19.1% improvements with MoCo on the severely biased datasets UCF101 and HMDB51, and 14.5% improvement on the less biased dataset Diving48.

Jinpeng Wang, Yuting Gao, Ke Li et al.

One significant factor we expect the video representation learning to capture, especially in contrast with the image representation learning, is the object motion. However, we found that in the current mainstream video datasets, some action categories are highly related with the scene where the action happens, making the model tend to degrade to a solution where only the scene information is encoded. For example, a trained model may predict a video as playing football simply because it sees the field, neglecting that the subject is dancing as a cheerleader on the field. This is against our original intention towards the video representation learning and may bring scene bias on different dataset that can not be ignored. In order to tackle this problem, we propose to decouple the scene and the motion (DSM) with two simple operations, so that the model attention towards the motion information is better paid. Specifically, we construct a positive clip and a negative clip for each video. Compared to the original video, the positive/negative is motion-untouched/broken but scene-broken/untouched by Spatial Local Disturbance and Temporal Local Disturbance. Our objective is to pull the positive closer while pushing the negative farther to the original clip in the latent space. In this way, the impact of the scene is weakened while the temporal sensitivity of the network is further enhanced. We conduct experiments on two tasks with various backbones and different pre-training datasets, and find that our method surpass the SOTA methods with a remarkable 8.1% and 8.8% improvement towards action recognition task on the UCF101 and HMDB51 datasets respectively using the same backbone.

Yexu Zhou, Yuting Gao, Yiran Huang et al.

An essential task in predictive maintenance is the prediction of the Remaining Useful Life (RUL) through the analysis of multivariate time series. Using the sliding window method, Convolutional Neural Network (CNN) and conventional Recurrent Neural Network (RNN) approaches have produced impressive results on this matter, due to their ability to learn optimized features. However, sequence information is only partially modeled by CNN approaches. Due to the flatten mechanism in conventional RNNs, like Long Short Term Memories (LSTM), the temporal information within the window is not fully preserved. To exploit the multi-level temporal information, many approaches are proposed which combine CNN and RNN models. In this work, we propose a new LSTM variant called embedded convolutional LSTM (ECLSTM). In ECLSTM a group of different 1D convolutions is embedded into the LSTM structure. Through this, the temporal information is preserved between and within windows. Since the hyper-parameters of models require careful tuning, we also propose an automated prediction framework based on the Bayesian optimization with hyperband optimizer, which allows for efficient optimization of the network architecture. Finally, we show the superiority of our proposed ECLSTM approach over the state-of-the-art approaches on several widely used benchmark data sets for RUL Estimation.

Yuting Gao, Zheng Huang, Yuchen Dai et al.

In this paper, we propose Double Supervised Network with Attention Mechanism (DSAN), a novel end-to-end trainable framework for scene text recognition. It incorporates one text attention module during feature extraction which enforces the model to focus on text regions and the whole framework is supervised by two branches. One supervision branch comes from context-level modelling and another comes from one extra supervision enhancement branch which aims at tackling inexplicit semantic information at character level. These two supervisions can benefit each other and yield better performance. The proposed approach can recognize text in arbitrary length and does not need any predefined lexicon. Our method outperforms the current state-of-the-art methods on three text recognition benchmarks: IIIT5K, ICDAR2013 and SVT reaching accuracy 88.6%, 92.3% and 84.1% respectively which suggests the effectiveness of the proposed method.

Yuchen Dai, Zheng Huang, Yuting Gao et al.

In this paper, we introduce a novel end-end framework for multi-oriented scene text detection from an instance-aware semantic segmentation perspective. We present Fused Text Segmentation Networks, which combine multi-level features during the feature extracting as text instance may rely on finer feature expression compared to general objects. It detects and segments the text instance jointly and simultaneously, leveraging merits from both semantic segmentation task and region proposal based object detection task. Not involving any extra pipelines, our approach surpasses the current state of the art on multi-oriented scene text detection benchmarks: ICDAR2015 Incidental Scene Text and MSRA-TD500 reaching Hmean 84.1% and 82.0% respectively. Morever, we report a baseline on total-text containing curved text which suggests effectiveness of the proposed approach.