Jiawei Liu, Xun Gong, Muli Yang, Xingrui Yu, Fen Fang, Xulei Yang, Ivor Tsang, Yunfeng hu, Hong Chen, Qing Guo. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Point-of-Interest (POI) recommendation aims to predict users' future locations based on their historical check-ins. Despite the success of recent deep learning approaches in capturing POI semantics and user behavior, they continue to face the persistent problem of data sparsity and incompleteness. In this paper, we introduce Multi-Objective Adversarial Imitation Recommender (MOAIR), a novel framework that integrates Generative Adversarial Imitation Learning with multi-objective to address this issue. MOAIR effectively captures user behavior patterns and spatial-temporal contextual information via graph-enhanced self-supervised state encoder and overcomes data sparsity by robustly learning from limited data and generating diverse samples. By accommodating diverse user patterns in the training data, the framework also mitigates the typical mode-collapse issue in generative adversarial learning and thus enhances the overall performance. MOAIR employs a multi-objective imitation learning architecture where the imitation learning agent (IL agent) explores the POI space and receives multifaceted reward signals. Utilizing the Paralleled Proximal Policy Optimization (3PO) framework to optimize multi-objectives, the IL agent ensures efficient and stable policy updates. Additionally, to address the issue of high noise in POI recommendation scenarios, we use a novel generative way to define our policy net and incorporate a variational bottleneck for regularization to enhance the stability of adversarial learning. Comprehensive experiments reveal the superior performance for MOAIR compared to other baseline approaches, especially with sparse training data.

Imitation learning is a central problem in reinforcement learning where the goal is to learn a policy that mimics the expert's behavior. In practice, it is often challenging to learn the expert policy from a limited number of demonstrations accurately due to the complexity of the state space. Moreover, it is essential to explore the environment and collect data to achieve beyond-expert performance. To overcome these challenges, we propose a novel imitation learning algorithm called Imitation Learning with Double Exploration (ILDE), which implements exploration in two aspects: (1) optimistic policy optimization via an exploration bonus that rewards state-action pairs with high uncertainty to potentially improve the convergence to the expert policy, and (2) curiosity-driven exploration of the states that deviate from the demonstration trajectories to potentially yield beyond-expert performance. Empirically, we demonstrate that ILDE outperforms the state-of-the-art imitation learning algorithms in terms of sample efficiency and achieves beyond-expert performance on Atari and MuJoCo tasks with fewer demonstrations than in previous work. We also provide a theoretical justification of ILDE as an uncertainty-regularized policy optimization method with optimistic exploration, leading to a regret growing sublinearly in the number of episodes.

Imitation learning (IL) has shown promise in various applications (e.g. robot locomotion) but is often limited to learning a single expert policy, constraining behavior diversity and robustness in unpredictable real-world scenarios. To address this, we introduce Quality Diversity Inverse Reinforcement Learning (QD-IRL), a novel framework that integrates quality-diversity optimization with IRL methods, enabling agents to learn diverse behaviors from limited demonstrations. This work introduces Extrinsic Behavioral Curiosity (EBC), which allows agents to receive additional curiosity rewards from an external critic based on how novel the behaviors are with respect to a large behavioral archive. To validate the effectiveness of EBC in exploring diverse locomotion behaviors, we evaluate our method on multiple robot locomotion tasks. EBC improves the performance of QD-IRL instances with GAIL, VAIL, and DiffAIL across all included environments by up to 185%, 42%, and 150%, even surpassing expert performance by 20% in Humanoid. Furthermore, we demonstrate that EBC is applicable to Gradient-Arborescence-based Quality Diversity Reinforcement Learning (QD-RL) algorithms, where it substantially improves performance and provides a generic technique for learning behavioral diverse policies. The source code of this work is provided at https://github.com/vanzll/EBC.

Metal defect detection is critical in industrial quality assurance, yet existing methods struggle with grayscale variations and complex defect states, limiting its robustness. To address these challenges, this paper proposes a Self-Adaptive Gamma Context-Aware SSM-based model(GCM-DET). Proposed detection framework integrating a Dynamic Gamma Correction (GC) module to enhance grayscale representation and optimize feature extraction for precise defect reconstruction. A State-Space Search Management (SSM) architecture captures robust multi-scale features, effectively handling defects of varying shapes and scales. Focal Loss is employed to mitigate class imbalance and refine detection accuracy. Additionally, the CD5-DET dataset is introduced, specifically designed for port container maintenance, featuring significant grayscale variations and intricate defect patterns. Experimental results demonstrate that the proposed model achieves substantial improvements, with [email protected] gains of 27.6%, 6.6%, and 2.6% on the CD5-DET, NEU-DET, and GC10-DET datasets. Dataset available at https://universe.roboflow.com/stanleysun233/cd5-det/dataset/1.

Incorrect boundary division, complex semantic representation, and differences in pronunciation and meaning often lead to errors in Chinese Named Entity Recognition (CNER). To address these issues, this paper proposes HREB-CRF framework: Hierarchical Reduced-bias EMA with CRF. The proposed method amplifies word boundaries and pools long text gradients through exponentially fixed-bias weighted average of local and global hierarchical attention. Experimental results on the MSRA, Resume, and Weibo datasets show excellent in F1, outperforming the baseline model by 1.1%, 1.6%, and 9.8%. The significant improvement in F1 shows evidences of strong effectiveness and robustness of approach in CNER tasks. Code available at https://github.com/StanleySun233/HREB-CRF.

Learning from imperfect demonstrations is a crucial challenge in imitation learning (IL). Unlike existing works that still rely on the enormous effort of expert demonstrators, we consider a more cost-effective option for obtaining a large number of demonstrations. That is, hire annotators to label actions for existing image records in realistic scenarios. However, action noise can occur when annotators are not domain experts or encounter confusing states. In this work, we introduce two particular forms of action noise, i.e., state-independent and state-dependent action noise. Previous IL methods fail to achieve expert-level performance when the demonstrations contain action noise, especially the state-dependent action noise. To mitigate the harmful effects of action noises, we propose a robust learning paradigm called USN (Uncertainty-aware Sample-selection with Negative learning). The model first estimates the predictive uncertainty for all demonstration data and then selects sampleswith high loss based on the uncertainty measures. Finally, it updates the model parameters with additional negative learning on the selected samples. Empirical results in Box2D tasks and Atari games show that USN consistently improves the final rewards of behavioral cloning, online imitation learning, and offline imitation learning methods under various action noises. The ratio of significant improvements is up to 94.44%. Moreover, our method scales to conditional imitation learning with real-world noisy commands in urban driving

Given data with noisy labels, over-parameterized deep networks can gradually memorize the data, and fit everything in the end. Although equipped with corrections for noisy labels, many learning methods in this area still suffer overfitting due to undesired memorization. In this paper, to relieve this issue, we propose stochastic integrated gradient underweighted ascent (SIGUA): in a mini-batch, we adopt gradient descent on good data as usual, and learning-rate-reduced gradient ascent on bad data; the proposal is a versatile approach where data goodness or badness is w.r.t. desired or undesired memorization given a base learning method. Technically, SIGUA pulls optimization back for generalization when their goals conflict with each other; philosophically, SIGUA shows forgetting undesired memorization can reinforce desired memorization. Experiments demonstrate that SIGUA successfully robustifies two typical base learning methods, so that their performance is often significantly improved.

Imitation learning in a high-dimensional environment is challenging. Most inverse reinforcement learning (IRL) methods fail to outperform the demonstrator in such a high-dimensional environment, e.g., Atari domain. To address this challenge, we propose a novel reward learning module to generate intrinsic reward signals via a generative model. Our generative method can perform better forward state transition and backward action encoding, which improves the module’s dynamics modeling ability in the environment. Thus, our module provides the imitation agent both the intrinsic intention of the demonstrator and a better exploration ability, which is critical for the agent to outperform the demonstrator. Empirical results show that our method outperforms state-of-the-art IRL methods on multiple Atari games, even with one-life demonstration. Remarkably, our method achieves performance that is up to 5 times the performance of the demonstration.

Learning with noisy labels is one of the hottest problems in weakly-supervised learning. Based on memorization effects of deep neural networks, training on small-loss instances becomes very promising for handling noisy labels. This fosters the state-of-the-art approach "Co-teaching" that cross-trains two deep neural networks using the small-loss trick. However, with the increase of epochs, two networks converge to a consensus and Co-teaching reduces to the self-training MentorNet. To tackle this issue, we propose a robust learning paradigm called Co-teaching+, which bridges the "Update by Disagreement” strategy with the original Co-teaching. First, two networks feed forward and predict all data, but keep prediction disagreement data only. Then, among such disagreement data, each network selects its small-loss data, but back propagates the small-loss data from its peer network and updates its own parameters. Empirical results on benchmark datasets demonstrate that Co-teaching+ is much superior to many state-of-the-art methods in the robustness of trained models.

We present a novel typhoon cloud prediction method via generative adversarial networks (GANs). Specifically, we develop a adversarial prediction model consisting of a generator and a discriminator. The generator generates the continuous future cloud images by learning the evolution trend of typhoon clouds from multiple continuous historical typhoon cloud images. In this way, the generator completes the visual predictions for typhoon clouds. On the other hand, the discriminator distinguishes the generated future cloud images from the real ones. Furthermore, we adopt a gradient difference loss function and a total variation loss function to improve the quality of the generated cloud images. The proposed method effectively predicts the whole spatial-temporal evolution of the typhoon clouds, resulting in a visual complement for classic typhoon prediction methods. The effectiveness of the proposed method has been evaluated in the real satellite cloud images.

Deep learning with noisy labels is practically challenging, as the capacity of deep models is so high that they can totally memorize these noisy labels sooner or later during training. Nonetheless, recent studies on the memorization effects of deep neural networks show that they would first memorize training data of clean labels and then those of noisy labels. Therefore in this paper, we propose a new deep learning paradigm called ''Co-teaching'' for combating with noisy labels. Namely, we train two deep neural networks simultaneously, and let them teach each other given every mini-batch: firstly, each network feeds forward all data and selects some data of possibly clean labels; secondly, two networks communicate with each other what data in this mini-batch should be used for training; finally, each network back propagates the data selected by its peer network and updates itself. Empirical results on noisy versions of MNIST, CIFAR-10 and CIFAR-100 demonstrate that Co-teaching is much superior to the state-of-the-art methods in the robustness of trained deep models.

We present a two-fold fusion framework that constructs comprehensive cost volumes for stereo matching. To this end, we develop fusion schemes at two key steps, i.e., a) the raw cost computation and b) the cost aggregation. Specifically, we commence by fusing both structure- and data-oriented features as raw costs. We then incorporate the guided filtered costs into the cross-based cost aggregation and obtain the fused aggregated costs. The fusion schemes at the two steps effectively complement each other and result in an accurate disparity map. Experiments on the Middlebury benchmark v3 demonstrate the state-of-the-art performance of our framework in terms of various metrics.

We develop an automatic oil spill segmentation method in terms of f-divergence minimization. We exploit f-divergence for measuring the disagreement between the distributions of ground-truth and generated oil spill segmentations. To render tractable optimization, we minimize the tight lower bound of the f-divergence by adversarial training a regressor and a generator, which are structured in different forms of deep neural networks separately. The generator aims at producing accurate oil spill segmentation, while the regressor characterizes discriminative distributions with respect to true and generated oil spill segmentations. It is the coplay between the generator net and the regressor net against each other that achieves a minimal of the maximum lower bound for the f-divergence. The adversarial strategy enhances the representational powers of both the generator and the regressor and avoids requesting large amounts of labeled data for training the deep network parameters. In addition, the trained generator net enables automatic oil spill detection that does not require manual initialization. Benefiting from the comprehensiveness of f-divergence for characterizing diversified distributions, our framework can accurately segment variously shaped oil spills in noisy synthetic aperture radar images. Experimental results validate the effectiveness of the proposed oil spill segmentation framework.

Segmentation of marine oil spill regions from SAR images has always been an important research topic in the remote sensing literature. One challenge for this problem is how to precisely preserve oil spill edges in segmentation results. To address this challenge, we present an edge sensitive algorithm for marine oil spill segmentation based on an energy minimization formulation. Specifically, we use an edge indicator in penalty terms of the energy function, for the purpose of detecting possible oil spill edges. Furthermore, a self-guided filtering scheme is incorporated into the energy function, which is capable of smoothing images without blurring edges. The energy minimization is conducted based on level set evolution, in which a double well distance regularization is involved for avoiding evolutionary irregularities. Empirical evaluations reveal that our method outperforms state of the art edge-based level set methods in marine oil spill segmentation.