With the burst of big data, 2D-3D cross-modal retrieval has received increasing attention, which aims to retrieve relevant data from one modality given the query from the other modality. In this paper, we study an underexplored yet practical problem of semi-supervised 2D-3D cross-modal retrieval, which could suffer from serious label scarcity in real-world applications. Moreover, the huge heterogeneous gap could deteriorate the process of learning from unlabeled data. In this work, we propose a novel approach named Decoupled Discriminative Learning with Bigraph-aware Alignment (DREAM) for semi-supervised 2D-3D cross-modal retrieval. The core of our DREAM is to decouple the label prediction and reliability measurement processes to reduce overconfident samples in discriminative learning. In particular, we enhance a label prediction module with label propagation from labeled samples and additionally introduce a reliability measurement module to learn the scores of predicted labels. To reduce class-related bias, we compare reliability scores with class-specific adaptive thresholds to identify samples for additional learning. In addition, negative labels are estimated for unselected samples, which guides soft semantic learning to make the best use of all the information. To further minimize the heterogeneous gap, we build a bigraph graph that connects cross-modal similar examples and then conduct learning to cluster with most edges kept for alignment. Extensive experiments on several benchmark datasets validate the superiority of the proposed DREAM.

Clinical diagnosis prediction models, when provided with a patient's medical history, aim to detect potential diseases early, facilitating timely intervention and improving prognostic outcomes. However, the inherent scarcity of patient data and large disease candidate space often pose challenges in developing satisfactory models for this intricate task. The exploration of leveraging Large Language Models (LLMs) for encapsulating clinical decision processes has been limited. We introduce MERA, a clinical diagnosis prediction model that bridges pertaining natural language knowledge with medical practice. We apply hierarchical contrastive learning on a disease candidate ranking list to alleviate the large decision space issue. With concept memorization through fine-tuning, we bridge the natural language clinical knowledge with medical codes. Experimental results on MIMIC-III and IV datasets show that MERA achieves the state-of-the-art diagnosis prediction performance and dramatically elevates the diagnosis prediction capabilities of generative LMs.

Yijia Xiao, Runhui Wang, Luyang Kong, Davor Golac, Wei Wang. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.

Recent studies reveal that large language models (LLMs) often struggle to resolve conflicting instructions embedded within hierarchical prompts, resulting in decreased compliance with system-level directives and compromising the reliability of safety-critical applications. While earlier approaches attempt to improve instruction hierarchy awareness through prompt engineering or embedding-level modifications, they typically lack structural modeling and either offer limited gains or require extensive fine-tuning. In this work, we introduce $\textbf{FocalLoRA}$, a parameter-efficient and structure-aware framework that strengthens hierarchical instruction adherence by selectively optimizing structurally critical attention heads, referred to as $\textit{focal heads}$, which exhibit heightened sensitivity to instruction conflicts. Experiments across multiple models and a dedicated benchmark demonstrate that FocalLoRA markedly enhances system instruction compliance with minimal tuning cost. For instance, on Llama-8B, fine-tuning only 0.0188\% of parameters yields a 35.52\% $\uparrow$ in system instruction compliance.

Hashing aims to learn a binary-output function that maps an image to a binary vector, which has received increasing attention with its potential in large-scale visual similarity search. Recently, supervised hashing methods have shown remarkable performance, but they assume that all examples are properly labeled. While in reality, it is unsurprising that we may encounter a range of label noise, which may significantly degrade retrieval performance. In response, we propose a noise-resistant Hashing Contrastive learning with hybrid selection (STAR). Specifically, STAR first develops noise-resistant hashing contrastive learning to preserve the similarity structure against label noise. In addition, we propose a hybrid sample selection strategy from the view of both Hamming distance and output uncertainty, which identifies reliable clean examples. Finally, to get rid of potential memorizing of noisy data, we incorporate both clean samples and noisy samples into selective centroid learning, which minimizes distances between clean samples and their centroids while pushing noisy samples away from negative centroids. Extensive experiments validate the efficacy of STAR.

Peer review is fundamental to the integrity and advancement of scientific publication.Traditional methods of peer review analyses often rely on exploration and statistics of existing peer review data, which do not adequately address the multivariate nature of the process, account for the latent variables, and are further constrained by privacy concerns due to the sensitive nature of the data.We introduce AGENTREVIEW, the first large language model (LLM) based peer review simulation framework, which effectively disentangles the impacts of multiple latent factors and addresses the privacy issue.Our study reveals significant insights, including a notable 37.1% variation in paper decisions due to reviewers' biases, supported by sociological theories such as the social influence theory, altruism fatigue, and authority bias.We believe that this study could offer valuable insights to improve the design of peer review mechanisms.Our code is available at https://github.com/Ahren09/AgentReview.

Yijia Xiao, Yiqiao Jin, Yushi Bai, Yue Wu, Xianjun Yang, Xiao Luo, Wenchao Yu, Xujiang Zhao, Yanchi Liu, Quanquan Gu, Haifeng Chen, Wei Wang, Wei Cheng. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. 2024.

As an important problem in searching system development, domain adaptive retrieval seeks to train a retrieval model with both labeled source samples and unlabeled target samples. Although several domain adaptive hashing algorithms have been proposed to handle the problem with high efficiency, they often presume that source and target domains share all classes. However, prior knowledge about the label space on the target domain is hard to obtain in reality. To tackle this, we study a novel and challenging problem of universal domain adaptive retrieval, which evidently increases the difficulty of effective domain alignment. In this paper, we propose a hashing method namedRelational and prOtotypicalStructure lEarning (ROSE) to solve the problem. In particular, to overcome domain shift, we construct a relational structure depicting cross-domain similar pairs based on ranking statistics, then learn from the structure by maximizing the similarity between similar pairs compared with challenging negatives. Moreover, target private samples are detected using the min-max criterion, which helps to construct hashing prototypes in the Hamming space. On this basis, we combine prototypical structure learning with online clustering in the Hamming space, which improves target semantic learning under label deficiency. Extensive experiments on several benchmarks demonstrate that our proposed ROSE significantly outperforms a wide range of state-of-the-art methods. Our source code is available athttps://github.com/WillDreamer/Rose.git.

Numerous benchmarks have been established to assess the performance of foundation models on open-ended question answering, which serves as a comprehensive test of a model's ability to understand and generate language in a manner similar to humans.Most of these works focus on proposing new datasets, however, we see two main issues within previous benchmarking pipelines, namely testing leakage and evaluation automation. In this paper, we propose a novel benchmarking framework, Language-Model-as-an-Examiner, where the LM serves as a knowledgeable examiner that formulates questions based on its knowledge and evaluates responses in a reference-free manner. Our framework allows for effortless extensibility as various LMs can be adopted as the examiner, and the questions can be constantly updated given more diverse trigger topics. For a more comprehensive and equitable evaluation, we devise three strategies: (1) We instruct the LM examiner to generate questions across a multitude of domains to probe for a broad acquisition, and raise follow-up questions to engage in a more in-depth assessment. (2) Upon evaluation, the examiner combines both scoring and ranking measurements, providing a reliable result as it aligns closely with human annotations. (3) We additionally propose a decentralized Peer-examination method to address the biases in a single examiner. Our data and benchmarking results are available at: http://lmexam.xlore.cn.

Drug resistance is a major threat to the global health and a significant concern throughout the clinical treatment of diseases and drug development. The mutation in proteins that is related to drug binding is a common cause for adaptive drug resistance. Therefore, quantitative estimations of how mutations would affect the interaction between a drug and the target protein would be of vital significance for the drug development and the clinical practice. Computational methods that rely on molecular dynamics simulations, Rosetta protocols, as well as machine learning methods have been proven to be capable of predicting ligand affinity changes upon protein mutation. However, the severely limited sample size and heavy noise induced overfitting and generalization issues have impeded wide adoption of machine learning for studying drug resistance. In this paper, we propose a robust machine learning method, termed SPLDExtraTrees, which can accurately predict ligand binding affinity changes upon protein mutation and identify resistance-causing mutations. Especially, the proposed method ranks training data following a specific scheme that starts with easy-to-learn samples and gradually incorporates harder and diverse samples into the training, and then iterates between sample weight recalculations and model updates. In addition, we calculate additional physics-based structural features to provide the machine learning model with the valuable domain knowledge on proteins for this data-limited predictive tasks. The experiments substantiate the capability of the proposed method for predicting kinase inhibitor resistance under three scenarios, and achieves predictive accuracy comparable to that of molecular dynamics and Rosetta methods with much less computational costs.