Accurate drug-drug interaction (DDI) prediction is crucial for optimizing the efficacy of combination therapies and minimizing adverse effects. Most existing methods rely on single features and struggle to integrate structural and sequential drug information. Additionally, prediction bias caused by class imbalance remains a significant challenge. To address these issues, this study proposes a multi-source example-driven learning framework for DDI (MEDL-DDI) that jointly models structural and sequential drug representations to achieve robust multimodal fusion and mitigate class imbalance. MEDL-DDI enriches SMILES with chemical knowledge, extracts global semantic features via a Transformer, and identifies key substructures through a graph information bottleneck. Moreover, an example-driven mechanism guided by example centers enhances the model's ability to recognize minority classes. Experimental results on three benchmark datasets validate that MEDL-DDI outperforms state-of-the-art methods. The case study on cardiovascular drug interactions further highlights MEDL-DDI's practical value and applicability.

Materials science text mining (MSTM), involving tasks like property extraction and synthesis action retrieval, is pivotal for advancing research by deriving critical insights from scientific literature. Descriptors, serving as essential task labels, often vary in meaning depending on researchers' usage purposes across different mining tasks. (e.g., 'Material' can refer to both synthesis components and participants in fuel cell experiment). This meaning difference makes it difficult for existing methods, fine-tuned to specific task, to handle the same descriptors in other tasks. To overcome above limitation, we propose MatDuck, a simple and effective approach for Zero-Shot MSTM by evoking material knowledge within Large Language Models (LLMs). Specifically, inspired by the Duck Typing principles in programming languages, we present a ClassDefinition-Style Descriptor generation method that evokes task-specific characteristics to address usage variation. Subsequently, we introduce code-style in-context learning for zero-shot tasks, reframing them into code to leverage LLMs' proficiency in code understanding. Extensive experiments on eight benchmark datasets demonstrate that MatDuck, as a plug-and-play approach, significantly improves the Zero-Shot MSTM performance of LLMs by an average of 11.3% across seven tasks.

In this paper, we propose a new multi-modal task, termed audio-visual instance segmentation (AVIS), which aims to simultaneously identify, segment and track individual sounding object instances in audible videos. To facilitate this research, we introduce a high-quality benchmark named AVISeg, containing over 90K instance masks from 26 semantic categories in 926 long videos. Additionally, we propose a strong baseline model for this task. Our model first localizes sound source within each frame, and condenses object-specific contexts into concise tokens. Then it builds long-range audio-visual dependencies between these tokens using window-based attention, and tracks sounding objects among the entire video sequences. Extensive experiments reveal that our method performs best on AVISeg, surpassing the existing methods from related tasks. We further conduct the evaluation on several multi-modal large models. Unfortunately, they exhibits subpar performance on instance-level sound source localization and temporal perception. We expect that AVIS will inspire the community towards a more comprehensive multi-modal understanding. Dataset and code is available at https://github.com/ruohaoguo/avis.

Molecular Relational Learning (MRL) is widely applied in molecular sciences. Recent studies attempt to retain molecular core information (e.g., substructures) by Graph Information Bottleneck but primarily focus on information compression without considering the causal dependencies of chemical reactions among substructures. This oversight neglects the core factors that determine molecular relationships, making maintaining stable MRL in distribution-shifted data challenging. To bridge this gap, we propose the Causal Subgraph Information Bottleneck (CausalGIB) for stable MRL. CausalGIB leverages causal dependency to guide substructure representation and integrates subgraph information bottleneck to optimize the core substructure representation, generating stable representations. Specifically, we distinguish causal and confounding substructures by noise injection and substructure interaction based on causal analysis. Furthermore, by minimizing the discrepancy between causal and confounding information within subgraph information bottleneck, CausalGIB captures core substructures composed of causal substructures and aggregates them into molecular representations to improve their stability. Experimental results on nine datasets demonstrate that CausalGIB outperforms state-of-the-art models in two tasks and significantly enhances model’s stability in distribution-shifted data.

Audio-Visual Video Parsing (AVVP) task aims to parse the event categories and occurrence times from audio and visual modalities in a given video. Existing methods usually focus on implicitly modeling audio and visual features through weak labels, without mining semantic relationships for different modalities and explicit modeling of event temporal dependencies. This makes it difficult for the model to accurately parse event information for each segment under weak supervision, especially when high similarity between segmental modal features leads to ambiguous event boundaries. Hence, we propose a multimodal optimization framework, TeMTG, that combines text enhancement and multi-hop temporal graph modeling. Specifically, we leverage pre-trained multimodal models to generate modality-specific text embeddings, and fuse them with audio-visual features to enhance the semantic representation of these features. In addition, we introduce a multi-hop temporal graph neural network, which explicitly models the local temporal relationships between segments, capturing the temporal continuity of both short-term and long-range events. Experimental results demonstrate that our proposed method achieves state-of-the-art (SOTA) performance in multiple key indicators in the LLP dataset.

RGB images provide a practical and cost-effective method of caries detection. However, the ambiguity of RGB caries images may lead to labeling errors during annotation, which can reduce the robustness of caries classification models. To address this, we propose Core Inter-Category Contrastive Learning (CICC) to improve the robustness of caries classification models. Rather than relying on traditional network fine-tuning, CICC focuses on improving the robustness of the model to label errors from a novel perspective by identifying core data that are highly relevant to the caries category. CICC utilizes the Jensen-Shannon Divergence to select core data, mitigating the impact of label errors on model performance. Inter-Category Contrastive Learning enhances feature representations of samples from different categories to improve the model's discrimination between caries categories. We validated the effectiveness of CICC in improving model robustness from model optimization and experimental results. Extensive experiments demonstrate that CICC significantly outperforms other comparative methods in caries classification performance and robustness. Our code is available at: https://github.com/papercode-for-cheung/CICC.

Abstract Large Language Models (LLMs) offer a promising alternative to traditional Materials Science Text Mining (MSTM) by reducing the need for extensive data labeling and fine-tuning. However, existing zero-/few-shot methods still face limitations in aligning with personalized needs in scientific discovery. To address this, we propose ClassMATe, an active knowledge structuring approach for MSTM. Specifically, we first propose a class definition stylization method to structure knowledge, enabling explicit clustering of latent material knowledge in LLMs for enhanced inference. To align with the scientists’ needs, we propose an active needs refining strategy that iteratively clarifies needs by learning from uncertainty-aware hard samples of LLMs, further refining the knowledge structuring. Extensive experiments on seven tasks and eight datasets show that ClassMATe, as a plug-and-play method, achieves performance comparable to supervised learning without requiring fine-tuning or extra knowledge base, highlighting the potential to bridge the gap between LLMs’ latent knowledge and real-world scientific applications.1

Exploring new therapeutic diseases for existing drugs plays an essential role in reducing drug development costs. However, existing methods for predicting drug–disease association (DDA) lack fusion to multi-neighborhood information, which limits their ability to generalize and forces them to rely on prior knowledge. To this end, we propose a novel DDA model called the Neighborhood Contrastive Learning Heterogeneous Networks (NCH-DDA). NCH-DDA uses both single-neighborhood and multi-neighborhood feature extraction modules to extract important features of drugs and diseases in parallel from multiple potential spaces, such as heterogeneous networks and similarity networks. NCH-DDA fuses single-neighborhood and multi-neighborhood features using contrastive learning to enhance information interaction in different neighborhood spaces, ultimately obtaining universal domain features of drugs and diseases. NCH-DDA uses a combination of predictive loss and triplet loss to reduce dependence on prior knowledge. In different partition schemes of multiple datasets, NCH-DDA achieved the best performance in predicting DDA, outperforming several current state-of-the-art methods. Moreover, NCH-DDA demonstrated better performance in experiments on data sparsity and drug repositioning for Alzheimer’s disease, indicating its greater potential in DDA prediction with sparse omics data and drug repositioning applications.

Drug substructure plays a crucial role in predicting drug-drug interaction (DDI) with combination drugs for disease therapies. In order to exploit the effect of drug substructure on DDI prediction, we propose a Key Substructure-aware Graph Transformer Network for Drug-drug Interaction Prediction (KSGTN-DDI). First, the substructure-adaptive graph Transformer module adaptively explicit encoding of drug structures information. Then, the key substructure-aware module calculates the importance of different substructures in DDI prediction. Finally, the calculated important substructure aggregation features are used to reconstruct the drug-drug interactions. Relevant experiments indicate that the performance of KSGTN-DDI outperforms other DDI prediction models.

BACKGROUND: The main task of medical entity disambiguation is to link mentions, such as diseases, drugs, or complications, to standard entities in the target knowledge base. To our knowledge, models based on Bidirectional Encoder Representations from Transformers (BERT) have achieved good results in this task. Unfortunately, these models only consider text in the current document, fail to capture dependencies with other documents, and lack sufficient mining of hidden information in contextual texts. RESULTS: We propose B-LBConA, which is based on Bio-LinkBERT and context-aware mechanism. Specifically, B-LBConA first utilizes Bio-LinkBERT, which is capable of learning cross-document dependencies, to obtain embedding representations of mentions and candidate entities. Then, cross-attention is used to capture the interaction information of mention-to-entity and entity-to-mention. Finally, B-LBConA incorporates disambiguation clues about the relevance between the mention context and candidate entities via the context-aware mechanism. CONCLUSIONS: Experiment results on three publicly available datasets, NCBI, ADR and ShARe/CLEF, show that B-LBConA achieves a signifcantly more accurate performance compared with existing models.