Enhancing Safety in Reinforcement Learning with Human Feedback via Rectified Policy Optimization

Balancing helpfulness and safety (harmlessness) is a critical challenge in aligning large language models (LLMs). Current approaches often decouple these two objectives, training separate preference models for helpfulness and safety, while framing safety as a constraint within a constrained Markov Decision Process (CMDP) framework. However, these methods can lead to ``safety interference'', where average-based safety constraints compromise the safety of some prompts in favor of others. To address this issue, we propose \textbf{Rectified Policy Optimization (RePO)}, which replaces the average safety constraint with stricter (per prompt) safety constraints. At the core of RePO is a policy update mechanism driven by rectified policy gradients, which penalizes the strict safety violation of every prompt, thereby enhancing safety across nearly all prompts. Our experiments on Alpaca-7B demonstrate that RePO improves the safety alignment and reduces the safety interference compared to baseline methods. Code is available at https://github.com/pxyWaterMoon/RePO.

DiffRetouch: Using Diffusion to Retouch on the Shoulder of Experts

Image retouching aims to enhance the visual quality of photos. Considering the different aesthetic preferences of users, the target of retouching is subjective. However, current retouching methods mostly adopt deterministic models, which not only neglects the style diversity in the expert-retouched results and tends to learn an average style during training, but also lacks sample diversity during inference. In this paper, we propose a diffusion-based method, named DiffRetouch. Thanks to the excellent distribution modeling ability of diffusion, our method can capture the complex fine-retouched distribution covering various visual-pleasing styles in the training data. Moreover, four image attributes are made adjustable to provide a user-friendly editing mechanism. By adjusting these attributes in specified ranges, users are allowed to customize preferred styles within the learned fine-retouched distribution. Additionally, the affine bilateral grid and contrastive learning scheme are introduced to handle the problem of texture distortion and control insensitivity respectively. Extensive experiments have demonstrated the superior performance of our method on visually appealing and sample diversity. The code will be made available to the community.

Diffusion-based Blind Text Image Super-Resolution

Recovering degraded low-resolution text images is challenging, especially for Chinese text images with complex strokes and severe degradation in real-world scenarios. Ensuring both text fidelity and style realness is crucial for high-quality text image super-resolution. Recently, diffusion models have achieved great success in natural image synthesis and restoration due to their powerful data distribution modeling abilities and data generation capabilities. In this work, we propose an Image Diffusion Model (IDM) to restore text images with realistic styles. For diffusion models, they are not only suitable for modeling realistic image distribution but also appropriate for learning text distribution. Since text prior is important to guarantee the correctness of the restored text structure according to existing arts, we also propose a Text Diffusion Model (TDM) for text recognition which can guide IDM to generate text images with correct structures. We further propose a Mixture of Multi-modality module (MoM) to make these two diffusion models cooperate with each other in all the diffusion steps. Extensive experiments on synthetic and real-world datasets demonstrate that our Diffusion-based Blind Text Image Super-Resolution (DiffTSR) can restore text images with more accurate text structures as well as more realistic appearances simultaneously.

Predicting Scores of Various Aesthetic Attribute Sets by Learning from Overall Score Labels

Now many mobile phones embed deep-learning models for evaluation or guidance on photography. These models cannot provide detailed results like human pose scores or scene color scores because of the rare of corresponding aesthetic attribute data. However, the annotation of image aesthetic attribute scores requires experienced artists and professional photographers, which hinders the collection of large-scale fully-annotated datasets. In this paper, we propose to replace image attribute labels with feature extractors. First, a novel aesthetic attribute evaluation framework based on attribute features is proposed to predict attribute scores and overall scores. We call it the F2S (attribute features to attribute scores) model. We use networks from different tasks to provide attribute features to our F2S models. Then, we define an aesthetic attribute contribution to describe the role of aesthetic attributes throughout an image and use it with the attribute scores and the overall scores to train our F2S model. Sufficient experiments on publicly available datasets demonstrate that our F2S model achieves comparable performance with those trained on the datasets with fully-annotated aesthetic attribute score labels. Our method makes it feasible to learn meaningful attribute scores for various aesthetic attribute sets in different types of images with only overall aesthetic scores.

Focusing on Persons: Colorizing Old Images Learning from Modern Historical Movies

In industry, there exist plenty of scenarios where old gray photos need to be automatically colored, such as video sites and archives. In this paper, we present the HistoryNet focusing on historical person's diverse high fidelity clothing colorization based on fine grained semantic understanding and prior. Colorization of historical persons is realistic and practical, however, existing methods do not perform well in the regards. In this paper, a HistoryNet including three parts, namely, classification, fine grained semantic parsing and colorization, is proposed. Classification sub-module supplies classifying of images according to the eras, nationalities and garment types; Parsing sub-network supplies the semantic for person contours, clothing and background in the image to achieve more accurate colorization of clothes and persons and prevent color overflow. In the training process, we integrate classification and semantic parsing features into the coloring generation network to improve colorization. Through the design of classification and parsing subnetwork, the accuracy of image colorization can be improved and the boundary of each part of image can be more clearly. Moreover, we also propose a novel Modern Historical Movies Dataset (MHMD) containing 1,353,166 images and 42 labels of eras, nationalities, and garment types for automatic colorization from 147 historical movies or TV series made in modern time. Various quantitative and qualitative comparisons demonstrate that our method outperforms the state-of-the-art colorization methods, especially on military uniforms, which has correct colors according to the historical literatures.

Learning Event-Based Motion Deblurring

Recovering sharp video sequence from a motion-blurred image is highly ill-posed due to the significant loss of motion information in the blurring process. For event-based cameras, however, fast motion can be captured as events at high time rate, raising new opportunities to exploring effective solutions. In this paper, we start from a sequential formulation of event-based motion deblurring, then show how its optimization can be unfolded with a novel end-to-end deep architecture. The proposed architecture is a convolutional recurrent neural network that integrates visual and temporal knowledge of both global and local scales in principled manner. To further improve the reconstruction, we propose a differentiable directional event filtering module to effectively extract rich boundary prior from the stream of events. We conduct extensive experiments on the synthetic GoPro dataset and a large newly introduced dataset captured by a DAVIS240C camera. The proposed approach achieves state-of-the-art reconstruction quality, and generalizes better to handling real-world motion blur.

Aesthetic Attributes Assessment of Images

Image aesthetic quality assessment has been a relatively hot topic during the last decade. Most recently, comments type assessment (aesthetic captions) has been proposed to describe the general aesthetic impression of an image using text. In this paper, we propose Aesthetic Attributes Assessment of Images, which means the aesthetic attributes captioning. This is a new formula of image aesthetic assessment, which predicts aesthetic attributes captions together with the aesthetic score of each attribute. We introduce a new dataset named \emph{DPC-Captions} which contains comments of up to 5 aesthetic attributes of one image through knowledge transfer from a full-annotated small-scale dataset. Then, we propose Aesthetic Multi-Attribute Network (AMAN), which is trained on a mixture of fully-annotated small-scale PCCD dataset and weakly-annotated large-scale DPC-Captions dataset. Our AMAN makes full use of transfer learning and attention model in a single framework. The experimental results on our DPC-Captions and PCCD dataset reveal that our method can predict captions of 5 aesthetic attributes together with numerical score assessment of each attribute. We use the evaluation criteria used in image captions to prove that our specially designed AMAN model outperforms traditional CNN-LSTM model and modern SCA-CNN model of image captions.