Peng Yi

Guohai Xu, Jiayi Liu, Ming Yan et al.

With the rapid evolution of large language models (LLMs), there is a growing concern that they may pose risks or have negative social impacts. Therefore, evaluation of human values alignment is becoming increasingly important. Previous work mainly focuses on assessing the performance of LLMs on certain knowledge and reasoning abilities, while neglecting the alignment to human values, especially in a Chinese context. In this paper, we present CValues, the first Chinese human values evaluation benchmark to measure the alignment ability of LLMs in terms of both safety and responsibility criteria. As a result, we have manually collected adversarial safety prompts across 10 scenarios and induced responsibility prompts from 8 domains by professional experts. To provide a comprehensive values evaluation of Chinese LLMs, we not only conduct human evaluation for reliable comparison, but also construct multi-choice prompts for automatic evaluation. Our findings suggest that while most Chinese LLMs perform well in terms of safety, there is considerable room for improvement in terms of responsibility. Moreover, both the automatic and human evaluation are important for assessing the human values alignment in different aspects. The benchmark and code is available on ModelScope and Github.

Xinlei Liu, Tao Hu, Jichao Xie et al.

Gradient-based attacks are important methods for evaluating model robustness. However, since the proposal of APGD, it has been difficult for such methods to achieve significant breakthroughs. To achieve such an effect, we first analyze the issue of "high-loss non-adversarial examples" that degrades attack performance in previous methods, and prove that this issue arises from inappropriate objectives for adversarial example generation. Subsequently, we reconstruct the objective as "maximizing the difference between the non-ground-truth label probability upper bound and the ground-truth label probability", and proposes a novel and powerful gradient-based attack method named Sequential Difference Maximization (SDM). SDM establishes a three-layer optimization framework of "cycle-stage-step". It adopts the negative probability loss function and the Directional Probability Difference Ratio (DPDR) loss function in the initial and subsequent optimization stages, respectively, and approaches the ideal objective of adversarial example generation via stage-wise sequential optimization. Experiments demonstrate that compared with previous state-of-the-art methods, SDM not only achieves stronger attack performance but also exhibits superior cost-effectiveness. The code is available at https://github.com/X-L-Liu/ICML-SDM.

Wenting Liu, Jinlong Lei, Peng Yi et al.

This paper considers no-regret learning for repeated continuous-kernel games with lossy bandit feedback. Since it is difficult to give the explicit model of the utility functions in dynamic environments, the players' action can only be learned with bandit feedback. Moreover, because of unreliable communication channels or privacy protection, the bandit feedback may be lost or dropped at random. Therefore, we study the asynchronous online learning strategy of the players to adaptively adjust the next actions for minimizing the long-term regret loss. The paper provides a novel no-regret learning algorithm, called Online Gradient Descent with lossy bandits (OGD-lb). We first give the regret analysis for concave games with differentiable and Lipschitz utilities. Then we show that the action profile converges to a Nash equilibrium with probability 1 when the game is also strictly monotone. We further provide the mean square convergence rate $\mathcal{O}\left(k^{-2\min\{β, 1/6\}}\right)$ when the game is $β-$ strongly monotone. In addition, we extend the algorithm to the case when the loss probability of the bandit feedback is unknown, and prove its almost sure convergence to Nash equilibrium for strictly monotone games. Finally, we take the resource management in fog computing as an application example, and carry out numerical experiments to empirically demonstrate the algorithm performance.

Peng Yi, Yang Cao, Xin Kang et al.

Deep learning-empowered semantic communication is regarded as a promising candidate for future 6G networks. Although existing semantic communication systems have achieved superior performance compared to traditional methods, the end-to-end architecture adopted by most semantic communication systems is regarded as a black box, leading to the lack of explainability. To tackle this issue, in this paper, a novel semantic communication system with a shared knowledge base is proposed for text transmissions. Specifically, a textual knowledge base constructed by inherently readable sentences is introduced into our system. With the aid of the shared knowledge base, the proposed system integrates the message and corresponding knowledge from the shared knowledge base to obtain the residual information, which enables the system to transmit fewer symbols without semantic performance degradation. In order to make the proposed system more reliable, the semantic self-information and the source entropy are mathematically defined based on the knowledge base. Furthermore, the knowledge base construction algorithm is developed based on a similarity-comparison method, in which a pre-configured threshold can be leveraged to control the size of the knowledge base. Moreover, the simulation results have demonstrated that the proposed approach outperforms existing baseline methods in terms of transmitted data size and sentence similarity.

Chi Wan, Yixin Cui, Jiatong Du et al.

End-to-end autonomous driving requires adaptive and robust handling of complex and diverse traffic environments. However, prevalent single-mode planning methods attempt to learn an overall policy while struggling to acquire diversified driving skills to handle diverse scenarios. Therefore, this paper proposes GEMINUS, a Mixture-of-Experts end-to-end autonomous driving framework featuring a Global Expert and a Scene-Adaptive Experts Group, equipped with a Dual-aware Router. Specifically, the Global Expert is trained on the overall dataset, possessing robust performance. The Scene-Adaptive Experts are trained on corresponding scene subsets, achieving adaptive performance. The Dual-aware Router simultaneously considers scenario-level features and routing uncertainty to dynamically activate expert modules. Through the effective coupling of the Global Expert and the Scene-Adaptive Experts Group via the Dual-aware Router, GEMINUS achieves both adaptability and robustness across diverse scenarios. GEMINUS outperforms existing methods in the Bench2Drive closed-loop benchmark and achieves state-of-the-art performance in Driving Score and Success Rate, even with only monocular vision input. The code is available at https://github.com/newbrains1/GEMINUS.

Xinlei Liu, Tao Hu, Peng Yi et al.

Efficient adversarial attack methods are critical for assessing the robustness of computer vision models. In this paper, we reconstruct the optimization objective for generating adversarial examples as "maximizing the difference between the non-true labels' probability upper bound and the true label's probability," and propose a gradient-based attack method termed Sequential Difference Maximization (SDM). SDM establishes a three-layer optimization framework of "cycle-stage-step." The processes between cycles and between iterative steps are respectively identical, while optimization stages differ in terms of loss functions: in the initial stage, the negative probability of the true label is used as the loss function to compress the solution space; in subsequent stages, we introduce the Directional Probability Difference Ratio (DPDR) loss function to gradually increase the non-true labels' probability upper bound by compressing the irrelevant labels' probabilities. Experiments demonstrate that compared with previous SOTA methods, SDM not only exhibits stronger attack performance but also achieves higher attack cost-effectiveness. Additionally, SDM can be combined with adversarial training methods to enhance their defensive effects. The code is available at https://github.com/X-L-Liu/SDM.

Kui Jiang, Zhongyuan Wang, Zheng Wang et al.

Low-light image enhancement aims to improve an image's visibility while keeping its visual naturalness. Different from existing methods tending to accomplish the relighting task directly by ignoring the fidelity and naturalness recovery, we investigate the intrinsic degradation and relight the low-light image while refining the details and color in two steps. Inspired by the color image formulation (diffuse illumination color plus environment illumination color), we first estimate the degradation from low-light inputs to simulate the distortion of environment illumination color, and then refine the content to recover the loss of diffuse illumination color. To this end, we propose a novel Degradation-to-Refinement Generation Network (DRGN). Its distinctive features can be summarized as 1) A novel two-step generation network for degradation learning and content refinement. It is not only superior to one-step methods, but also capable of synthesizing sufficient paired samples to benefit the model training; 2) A multi-resolution fusion network to represent the target information (degradation or contents) in a multi-scale cooperative manner, which is more effective to address the complex unmixing problems. Extensive experiments on both the enhancement task and joint detection task have verified the effectiveness and efficiency of our proposed method, surpassing the SOTA by \textit{0.70dB on average and 3.18\% in mAP}, respectively. The code is available at \url{https://github.com/kuijiang0802/DRGN}.

Kui Jiang, Zhongyuan Wang, Peng Yi et al.

Rain streaks in the air appear in various blurring degrees and resolutions due to different distances from their positions to the camera. Similar rain patterns are visible in a rain image as well as its multi-scale (or multi-resolution) versions, which makes it possible to exploit such complementary information for rain streak representation. In this work, we explore the multi-scale collaborative representation for rain streaks from the perspective of input image scales and hierarchical deep features in a unified framework, termed multi-scale progressive fusion network (MSPFN) for single image rain streak removal. For similar rain streaks at different positions, we employ recurrent calculation to capture the global texture, thus allowing to explore the complementary and redundant information at the spatial dimension to characterize target rain streaks. Besides, we construct multi-scale pyramid structure, and further introduce the attention mechanism to guide the fine fusion of this correlated information from different scales. This multi-scale progressive fusion strategy not only promotes the cooperative representation, but also boosts the end-to-end training. Our proposed method is extensively evaluated on several benchmark datasets and achieves state-of-the-art results. Moreover, we conduct experiments on joint deraining, detection, and segmentation tasks, and inspire a new research direction of vision task-driven image deraining. The source code is available at \url{https://github.com/kuihua/MSPFN}.

Zhongyuan Wang, Guangcheng Wang, Baojin Huang et al.

In order to effectively prevent the spread of COVID-19 virus, almost everyone wears a mask during coronavirus epidemic. This almost makes conventional facial recognition technology ineffective in many cases, such as community access control, face access control, facial attendance, facial security checks at train stations, etc. Therefore, it is very urgent to improve the recognition performance of the existing face recognition technology on the masked faces. Most current advanced face recognition approaches are designed based on deep learning, which depend on a large number of face samples. However, at present, there are no publicly available masked face recognition datasets. To this end, this work proposes three types of masked face datasets, including Masked Face Detection Dataset (MFDD), Real-world Masked Face Recognition Dataset (RMFRD) and Simulated Masked Face Recognition Dataset (SMFRD). Among them, to the best of our knowledge, RMFRD is currently theworld's largest real-world masked face dataset. These datasets are freely available to industry and academia, based on which various applications on masked faces can be developed. The multi-granularity masked face recognition model we developed achieves 95% accuracy, exceeding the results reported by the industry. Our datasets are available at: https://github.com/X-zhangyang/Real-World-Masked-Face-Dataset.

Yang Lv, Jinlong Lei, Peng Yi

In this paper, we explore how to optimize task allocation for robot swarms in dynamic environments, emphasizing the necessity of formulating robust, flexible, and scalable strategies for robot cooperation. We introduce a novel framework using a decentralized partially observable Markov decision process (Dec_POMDP), specifically designed for distributed robot swarm networks. At the core of our methodology is the Local Information Aggregation Multi-Agent Deep Deterministic Policy Gradient (LIA_MADDPG) algorithm, which merges centralized training with distributed execution (CTDE). During the centralized training phase, a local information aggregation (LIA) module is meticulously designed to gather critical data from neighboring robots, enhancing decision-making efficiency. In the distributed execution phase, a strategy improvement method is proposed to dynamically adjust task allocation based on changing and partially observable environmental conditions. Our empirical evaluations show that the LIA module can be seamlessly integrated into various CTDE-based MARL methods, significantly enhancing their performance. Additionally, by comparing LIA_MADDPG with six conventional reinforcement learning algorithms and a heuristic algorithm, we demonstrate its superior scalability, rapid adaptation to environmental changes, and ability to maintain both stability and convergence speed. These results underscore LIA_MADDPG's outstanding performance and its potential to significantly improve dynamic task allocation in robot swarms through enhanced local collaboration and adaptive strategy execution.

Dixiao Wei, Peng Yi, Jinlong Lei et al.

Reinforcement Learning (RL) has demonstrated excellent decision-making potential in platoon coordination problems. However, due to the variability of coordination goals, the complexity of the decision problem, and the time-consumption of trial-and-error in manual design, finding a well performance reward function to guide RL training to solve complex platoon coordination problems remains challenging. In this paper, we formally define the Platoon Coordination Reward Design Problem (PCRDP), extending the RL-based cooperative platoon coordination problem to incorporate automated reward function generation. To address PCRDP, we propose a Large Language Model (LLM)-based Platoon coordination Reward Design (PCRD) framework, which systematically automates reward function discovery through LLM-driven initialization and iterative optimization. In this method, LLM first initializes reward functions based on environment code and task requirements with an Analysis and Initial Reward (AIR) module, and then iteratively optimizes them based on training feedback with an evolutionary module. The AIR module guides LLM to deepen their understanding of code and tasks through a chain of thought, effectively mitigating hallucination risks in code generation. The evolutionary module fine-tunes and reconstructs the reward function, achieving a balance between exploration diversity and convergence stability for training. To validate our approach, we establish six challenging coordination scenarios with varying complexity levels within the Yangtze River Delta transportation network simulation. Comparative experimental results demonstrate that RL agents utilizing PCRD-generated reward functions consistently outperform human-engineered reward functions, achieving an average of 10\% higher performance metrics in all scenarios.

Yang Lv, Jinlong Lei, Peng Yi

Two-stage Colonel Blotto game represents a typical adversarial resource allocation problem, in which two opposing agents sequentially allocate resources in a network topology across two phases: an initial resource deployment followed by multiple rounds of dynamic reallocation adjustments. The sequential dependency between game stages and the complex constraints imposed by the graph topology make it difficult for traditional approaches to attain a globally optimal strategy. To address these challenges, we propose a hierarchical graph Transformer framework called HGformer. By incorporating an enhanced graph Transformer encoder with structural biases and a two-agent hierarchical decision model, our approach enables efficient policy generation in large-scale adversarial environments. Moreover, we design a layer-by-layer feedback reinforcement learning algorithm that feeds the long-term returns from lower-level decisions back into the optimization of the higher-level strategy, thus bridging the coordination gap between the two decision-making stages. Experimental results demonstrate that, compared to existing hierarchical decision-making or graph neural network methods, HGformer significantly improves resource allocation efficiency and adversarial payoff, achieving superior overall performance in complex dynamic game scenarios.

Dixiao Wei, Peng Yi, Jinlong Lei et al.

Truck platooning technology enables a group of trucks to travel closely together, with which the platoon can save fuel, improve traffic flow efficiency, and improve safety. In this paper, we consider the platoon coordination problem in a large-scale transportation network, to promote cooperation among trucks and optimize the overall efficiency. Involving the regulation of both speed and departure times at hubs, we formulate the coordination problem as a complicated dynamic stochastic integer programming under network and information constraints. To get an autonomous, distributed, and robust platoon coordination policy, we formulate the problem into a model of the Decentralized-Partial Observable Markov Decision Process. Then, we propose a Multi-Agent Deep Reinforcement Learning framework named Trcuk Attention-QMIX (TA-QMIX) to train an efficient online decision policy. TA-QMIX utilizes the attention mechanism to enhance the representation of truck fuel gains and delay times, and provides explicit truck cooperation information during the training process, promoting trucks' willingness to cooperate. The training framework adopts centralized training and distributed execution, thus training a policy for trucks to make decisions online using only nearby information. Hence, the policy can be autonomously executed on a large-scale network. Finally, we perform comparison experiments and ablation experiments in the transportation network of the Yangtze River Delta region in China to verify the effectiveness of the proposed framework. In a repeated comparative experiment with 5,000 trucks, our method average saves 19.17\% of fuel with an average delay of only 9.57 minutes per truck and a decision time of 0.001 seconds.

Peng Yi, Kecheng Chen, Zhaoqi Ma et al.

Electroencephalogram (EEG) has shown a useful approach to produce a brain-computer interface (BCI). One-dimensional (1-D) EEG signal is yet easily disturbed by certain artifacts (a.k.a. noise) due to the high temporal resolution. Thus, it is crucial to remove the noise in received EEG signal. Recently, deep learning-based EEG signal denoising approaches have achieved impressive performance compared with traditional ones. It is well known that the characteristics of self-similarity (including non-local and local ones) of data (e.g., natural images and time-domain signals) are widely leveraged for denoising. However, existing deep learning-based EEG signal denoising methods ignore either the non-local self-similarity (e.g., 1-D convolutional neural network) or local one (e.g., fully connected network and recurrent neural network). To address this issue, we propose a novel 1-D EEG signal denoising network with 2-D transformer, namely EEGDnet. Specifically, we comprehensively take into account the non-local and local self-similarity of EEG signal through the transformer module. By fusing non-local self-similarity in self-attention blocks and local self-similarity in feed forward blocks, the negative impact caused by noises and outliers can be reduced significantly. Extensive experiments show that, compared with other state-of-the-art models, EEGDnet achieves much better performance in terms of both quantitative and qualitative metrics.

Peng Yi, Zhongyuan Wang, Kui Jiang et al.

Most recent video super-resolution (SR) methods either adopt an iterative manner to deal with low-resolution (LR) frames from a temporally sliding window, or leverage the previously estimated SR output to help reconstruct the current frame recurrently. A few studies try to combine these two structures to form a hybrid framework but have failed to give full play to it. In this paper, we propose an omniscient framework to not only utilize the preceding SR output, but also leverage the SR outputs from the present and future. The omniscient framework is more generic because the iterative, recurrent and hybrid frameworks can be regarded as its special cases. The proposed omniscient framework enables a generator to behave better than its counterparts under other frameworks. Abundant experiments on public datasets show that our method is superior to the state-of-the-art methods in objective metrics, subjective visual effects and complexity. Our code will be made public.