Shuai Niu, Jing Ma, Hongzhan Lin, Liang Bai, Zhihua Wang, Richard Yi Da Xu, Yunya Song, Xian Yang. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Treatment effect estimation from observational data is a fundamental problem in causal inference, and its critical challenge is to address the confounding bias arising from the confounders. The effectiveness of the conventional methods proposed to solve this problem depends on the unconfoundedness assumption. In practice, however, the unconfoundedness assumption is frequently violated since we cannot guarantee that all the confounders are measured. To this end, recent studies suggest using auxiliary network architectures to mine information about unmeasured confounders in the data to relax this assumption. However, these methods cannot address the confounding bias from unmeasured confounders unrelated to the network information. Inspired by the insight that some neighboring features that influence one's treatment choice (e.g., which movie to watch) but do not affect the outcome (e.g., assessment of the movie) can be treated as instrumental variables (IVs), we propose a novel Network Instrumental Variable Regression (NetIV) framework exploits IV information from neighborhoods to perform a two-stage regression for treatment effect estimation. Extensive experiments demonstrate that our NetIV method outperforms the state-of-the-art methods for treatment effect estimation in the presence of unmeasured confounders.

Colorectal cancer (CRC) is a significant global health concern, and early detection through screening plays a critical role in reducing mortality. While deep learning models have shown promise in improving polyp detection, classification, and segmentation, their generalization across diverse clinical environments, particularly with out-of-distribution (OOD) data, remains a challenge. Multi-center datasets like PolypGen have been developed to address these issues, but their collection is costly and time-consuming. Traditional data augmentation techniques provide limited variability, failing to capture the complexity of medical images. Diffusion models have emerged as a promising solution for generating synthetic polyp images, but the image generation process in current models mainly relies on segmentation masks as the condition, limiting their ability to capture the full clinical context. To overcome these limitations, we propose a Progressive Spectrum Diffusion Model (PSDM) that integrates diverse clinical annotations-such as segmentation masks, bounding boxes, and colonoscopy reports-by transforming them into compositional prompts. These prompts are organized into coarse and fine components, allowing the model to capture both broad spatial structures and fine details, generating clinically accurate synthetic images. By augmenting training data with PSDM-generated samples, our model significantly improves polyp detection, classification, and segmentation. For instance, on the PolypGen dataset, PSDM increases the F1 score by 2.12% and the mean average precision by 3.09%, demonstrating superior performance in OOD scenarios and enhanced generalization.

Recommender systems primarily utilize user–item interactions (i.e., collaborative information) and auxiliary textual information (i.e., semantic information) to infer user preferences and provide recommendations. With the advancement in large language models (LLMs), attempts have been made to incorporate their remarkable language comprehension capabilities into recommendation tasks. However, existing LLM4Rec methods face challenges in seamlessly integrating both collaborative and semantic information, as there is an inherent gap between these two types of data. Moreover, these methods struggle to capture the fine-grained distinctions in user preferences, which are essential in recommendation tasks, due to the loss design of LLMs. To address these issues, we propose a multi-stage prompt-tuning method for leveraging pre-trained LLMs in various recommendation tasks, named SCRec. Specifically, SCRec leverages Semantic and Collaborative information as supervision signals in two distinct stages: the semantic prompt-tuning stage and the collaborative prompt-tuning stage. This method breaks down user and item representations into semantic and collaborative perspectives, enabling a pre-trained LLM to first deduce the qualitative preferences of users over items from semantic information, and then generate quantitative recommendations from collaborative information. In addition, we propose a meta-mapping approach to provide personalized mapping functions for encoding collaborative information and integrate a novel numeric-informed head based on MSE loss for LLM in the second stage, which helps to better capture fine-grained distinctions in user preferences. Experiments on three public datasets for rating prediction and top-N recommendation tasks demonstrate that our method surpasses both conventional and LLM-based techniques, showing the strength of sequentially merging semantic and collaborative information in recommendation tasks.

Machine learning-assisted retrosynthesis prediction models have been gaining widespread adoption, though their performances oftentimes degrade significantly when deployed in real-world applications embracing out-of-distribution (OOD) molecules or reactions. Despite steady progress on standard benchmarks, our understanding of existing retrosynthesis prediction models under the premise of distribution shifts remains stagnant. To this end, we first formally sort out two types of distribution shifts in retrosynthesis prediction and construct two groups of benchmark datasets. Next, through comprehensive experiments, we systematically compare state-of-the-art retrosynthesis prediction models on the two groups of benchmarks, revealing the limitations of previous in-distribution evaluation and re-examining the advantages of each model. More remarkably, we are motivated by the above empirical insights to propose two model-agnostic techniques that can improve the OOD generalization of arbitrary off-the-shelf retrosynthesis prediction algorithms. Our preliminary experiments show their high potential with an average performance improvement of 4.6%, and the established benchmarks serve as a foothold for further retrosynthesis prediction research towards OOD generalization.

Retrosynthetic planning is a sequential decision-making process of identifying synthetic routes from the available building block materials to reach a desired target molecule. Though existing planning approaches show promisingly high solving rates and low costs, the trivial route cost evaluation via pre-trained forward reaction prediction models certainly falls short of real-world chemical practice. An alternative option is to annotate the actual cost of a route, such as yield, through chemical experiments or input from chemists, while this often leads to substantial query costs. In order to strike the balance between query costs and route quality evaluation, we propose an Active Retrosynthetic Planning (ARP) framework that remains compatible with the established retrosynthetic planners. On one hand, the proposed ARP trains an actor that decides whether to query the cost of a reaction; on the other hand, it resorts to a critic to estimate the value of a molecule with its preceding reaction cost as input. Those molecules with low reaction costs are preferred to expand first. We apply our framework to different existing approaches on both the benchmark and an expert dataset and demonstrate that it outperforms the existing state-of-the-art approach by 6.2\% in route quality while reducing the query cost by 12.8\%. In addition, ARP consistently plans high-quality routes with either abundant or sparse annotations.

Retrosynthesis aided by artificial intelligence has been a very active and bourgeoning area of research, for its critical role in drug discovery as well as material science. Three categories of solutions, i.e., template-based, template-free, and semi-template methods, constitute mainstream solutions to this problem. In this paper, we focus on template-free methods which are known to be less bothered by the template generalization issue and the atom mapping challenge. Among several remaining problems regarding template-free methods, failing to conform to chemical rules is pronounced. To address the issue, we seek for a pre-training solution to empower the pre-trained model with chemical rules encoded. Concretely, we enforce the atom conservation rule via a molecule reconstruction pre-training task, and the reaction rule that dictates reaction centers via a reaction type guided contrastive pre-training task. In our empirical evaluation, the proposed pre-training solution substantially improves the single-step retrosynthesis accuracies in three downstream datasets.

The advent of the big data era brought new opportunities and challenges to draw treatment effect in data fusion, that is, a mixed dataset collected from multiple sources (each source with an independent treatment assignment mechanism). Due to possibly omitted source labels and unmeasured confounders, traditional methods cannot estimate individual treatment assignment probability and infer treatment effect effectively. Therefore, we propose to reconstruct the source label and model it as a Group Instrumental Variable (GIV) to implement IV-based Regression for treatment effect estimation. In this paper, we conceptualize this line of thought and develop a unified framework (Meta-EM) to (1) map the raw data into a representation space to construct Linear Mixed Models for the assigned treatment variable; (2) estimate the distribution differences and model the GIV for the different treatment assignment mechanisms; and (3) adopt an alternating training strategy to iteratively optimize the representations and the joint distribution to model GIV for IV regression. Empirical results demonstrate the advantages of our Meta-EM compared with state-of-the-art methods. The project page with the code and the Supplementary materials is available at https://github.com/causal-machine-learning-lab/meta-em.

Network embedding is a process of encoding nodes into latent vectors by preserving network structure and content information. It is used in various applications, especially in recommender systems. In a social network setting, when recommending new friends to a user, the similarity between the user's embedding and the target friend will be examined. Traditional methods generate user node embedding without considering the recommendation target. No matter which target is to be recommended, the same embedding vector is generated for that particular user. This approach has its limitations. For example, a user can be both a computer scientist and a musician. When recommending music friends with potentially the same taste to him, we are interested in getting his representation that is useful in recommending music friends rather than computer scientists. His corresponding embedding should consider the user's musical features rather than those associated with computer science with the awareness that the recommendation targets are music friends. In order to address this issue, we propose a new framework which we name it as Recommendation Target-Aware Network embedding method (RTANet). Herein, the embedding of each user is no longer fixed to a constant vector, but it can vary according to their specific recommendation target. Concretely, RTANet assigns different attention weights to each neighbour node, allowing us to obtain the user's context information aggregated from its neighbours before transforming this context into its embedding. Different from other graph attention approaches, the attention weights in our work measure the similarity between each user's neighbour node and the target node, which in return generates the target-aware embedding. To demonstrate the effectiveness of our method, we compared RTANet with several state-of-the-art network embedding methods on four real-world datasets and showed that RTANet outperforms other comparative methods in the recommendation tasks.

The focus of our work is on diagnostic tasks in personalized learning, such as cognitive diagnosis and knowledge tracing. The goal of these tasks is to assess students' latent proficiency on knowledge concepts through analyzing their historical learning records. However, existing research has been limited to single-course scenarios; cross-course studies have not been explored due to a lack of dataset. We address this issue by constructing PTADisc, a Diverse, Immense, Student-centered dataset that emphasizes its sufficient Cross-course information for personalized learning. PTADisc includes 74 courses, 1,530,100 students, 4,054 concepts, 225,615 problems, and over 680 million student response logs. Based on PTADisc, we developed a model-agnostic Cross-Course Learner Modeling Framework (CCLMF) which utilizes relationships between students' proficiency across courses to alleviate the difficulty of diagnosing student knowledge state in cold-start scenarios. CCLMF uses a meta network to generate personalized mapping functions between courses. The experimental results on PTADisc verify the effectiveness of CCLMF with an average improvement of 4.2% on AUC. We also report the performance of baseline models for cognitive diagnosis and knowledge tracing over PTADisc, demonstrating that our dataset supports a wide scope of research in personalized learning. Additionally, PTADisc contains valuable programming logs and student-group information that are worth exploring in the future.

Driven by recent advances in neural networks, various Deep Embedding Clustering (DEC) based short text clustering models are being developed. In these works, latent representation learning and text clustering are performed simultaneously. Although these methods are becoming increasingly popular, they use pure cluster-oriented objectives, which can produce meaningless representations. To alleviate this problem, several improvements have been developed to introduce additional learning objectives in the clustering process, such as models based on contrastive learning. However, existing efforts rely heavily on learning meaningful representations at the instance level. They have limited focus on learning global representations, which are necessary to capture the overall data structure at the cluster level. In this paper, we propose a novel DEC model, which we named the deep embedded clustering model with cluster-level representation learning (DECCRL) to jointly learn cluster and instance level representations. Here, we extend the embedded topic modelling approach to introduce reconstruction constraints to help learn cluster-level representations. Experimental results on real-world short text datasets demonstrate that our model produces meaningful clusters.

Meta-learning has emerged as a potent paradigm for quick learning of few-shot tasks, by leveraging the meta-knowledge learned from meta-training tasks. Well-generalized meta-knowledge that facilitates fast adaptation in each task is preferred; however, recent evidence suggests the undesirable memorization effect where the meta-knowledge simply memorizing all meta-training tasks discourages task-specific adaptation and poorly generalizes. There have been several solutions to mitigating the effect, including both regularizer-based and augmentation-based methods, while a systematic understanding of these methods in a single framework is still lacking. In this paper, we offer a novel causal perspective of meta-learning. Through the lens of causality, we conclude the universal label space as a confounder to be the causing factor of memorization and frame the two lines of prevailing methods as different deconfounder approaches. Remarkably, derived from the causal inference principle of front-door adjustment, we propose two frustratingly easy but effective deconfounder algorithms, i.e., sampling multiple versions of the meta-knowledge via Dropout and grouping the meta-knowledge into multiple bins. The proposed causal perspective not only brings in the two deconfounder algorithms that surpass previous works in four benchmark datasets towards combating memorization, but also opens a promising direction for meta-learning.

The success of deep learning is partly attributed to the availability of massive data downloaded freely from the Internet. However, it also means that users' private data may be collected by commercial organizations without consent and used to train their models. Therefore, it's important and necessary to develop a method or tool to prevent unauthorized data exploitation. In this paper, we propose ConfounderGAN, a generative adversarial network (GAN) that can make personal image data unlearnable to protect the data privacy of its owners. Specifically, the noise produced by the generator for each image has the confounder property. It can build spurious correlations between images and labels, so that the model cannot learn the correct mapping from images to labels in this noise-added dataset. Meanwhile, the discriminator is used to ensure that the generated noise is small and imperceptible, thereby remaining the normal utility of the encrypted image for humans. The experiments are conducted in six image classification datasets, including three natural object datasets and three medical datasets. The results demonstrate that our method not only outperforms state-of-the-art methods in standard settings, but can also be applied to fast encryption scenarios. Moreover, we show a series of transferability and stability experiments to further illustrate the effectiveness and superiority of our method.