Radiology report generation involves translating visual signals from pixels into precise clinical language. Existing encoder-decoder models often suffer from hallucinations, generating plausible but incorrect medical findings. We propose GraphRAG-Rad, a novel architecture that integrates biomedical knowledge through a novel Latent Visual-Semantic Retrieval (VSR). Unlike traditional Retrieval-Augmented Generation (RAG) methods that rely on textual queries, our approach aligns visual embeddings with the latent space of the Knowledge Graph, PrimeKG. The retrieved sub-graph guides the Visual Encoder and the Multi-Hop Reasoning Module. The reasoning module simulates clinical deduction paths (Ground-Glass Opacity → Viral Pneumonia → COVID-19) before it combines the information with visual features in a Graph-Gated Cross-Modal Decoder. Experiments on the COV-CTR dataset demonstrate that GraphRAG-Rad achieves competitive performance with strong results across multiple metrics. Furthermore, ablation studies show that integrating latent retrieval and reasoning improves performance significantly compared to a visual-only baseline. Qualitative analysis further reveals interpretable attention maps. These maps explicitly link visual regions to symbolic medical concepts, effectively bridging the modality gap between vision and language.

There is a huge demand for information about climate change across all sectors as societies seek to mitigate and adapt to its impacts. However, the volume and complexity of climate information, which takes many formats including numerical, text, and tabular data, can make good information hard to access. Here we use Large Language Models (LLMs) and Retrieval Augmented Generation (RAG) to create an AI agent that provides accurate and complete information from the United Kingdom Climate Projections 2018 (UKCP18) data archive. To overcome the problematic hallucinations associated with LLMs, four phases of experiments were performed to optimize different components of our RAG framework, combining various recent retrieval strategies. Performance was evaluated using three statistical metrics (faithfulness, relevance, coverage) as well as human evaluation by subject matter experts. Results show that the best model significantly outperforms a generic LLM (GPT-3.5) and has high-quality outputs with positive ratings by human experts. The UKCP Chatbot developed here will enable access at scale to the UKCP18 climate archives, offering an important case study of using RAG-based LLM systems to communicate climate information.

Investigating whether pre-trained language models (LMs) can function as knowledge bases (KBs) has raised wide research interests recently. However, existing works focus on simple, triple-based, relational KBs, but omit more sophisticated, logic-based, conceptualised KBs such as OWL ontologies. To investigate an LM’s knowledge of ontologies, we propose OntoLAMA, a set of inference-based probing tasks and datasets from ontology subsumption axioms involving both atomic and complex concepts. We conduct extensive experiments on ontologies of different domains and scales, and our results demonstrate that LMs encode relatively less background knowledge of Subsumption Inference (SI) than traditional Natural Language Inference (NLI) but can improve on SI significantly when a small number of samples are given. We will open-source our code and datasets.

Medical document coding is the process of assigning labels from a structured label space (ontology – e.g., ICD-9) to medical documents. This process is laborious, costly, and error-prone. In recent years, efforts have been made to automate this process with neural models. The label spaces are large (in the order of thousands of labels) and follow a big-head long-tail label distribution, giving rise to few-shot and zero-shot scenarios. Previous efforts tried to address these scenarios within the model, leading to improvements on rare labels, but worse results on frequent ones. We propose data augmentation and synthesis techniques in order to address these scenarios. We further introduce an analysis technique for this setting inspired by confusion matrices. This analysis technique points to the positive impact of data augmentation and synthesis, but also highlights more general issues of confusion within families of codes, and underprediction.

Large-Scale Multi-Label Text Classification (LMTC) includes tasks with hierarchical label spaces, such as automatic assignment of ICD-9 codes to discharge summaries. Performance of models in prior art is evaluated with standard precision, recall, and F1 measures without regard for the rich hierarchical structure. In this work we argue for hierarchical evaluation of the predictions of neural LMTC models. With the example of the ICD-9 ontology we describe a structural issue in the representation of the structured label space in prior art, and propose an alternative representation based on the depth of the ontology. We propose a set of metrics for hierarchical evaluation using the depth-based representation. We compare the evaluation scores from the proposed metrics with previously used metrics on prior art LMTC models for ICD-9 coding in MIMIC-III. We also propose further avenues of research involving the proposed ontological representation.

We propose a novel attention network for document annotation with user-generated tags. The network is designed according to the human reading and annotation behaviour. Usually, users try to digest the title and obtain a rough idea about the topic first, and then read the content of the document. Present research shows that the title metadata could largely affect the social annotation. To better utilise this information, we design a framework that separates the title from the content of a document and apply a title-guided attention mechanism over each sentence in the content. We also propose two semantic-based loss regularisers that enforce the output of the network to conform to label semantics, i.e. similarity and subsumption. We analyse each part of the proposed system with two real-world open datasets on publication and question annotation. The integrated approach, Joint Multi-label Attention Network (JMAN), significantly outperformed the Bidirectional Gated Recurrent Unit (Bi-GRU) by around 13%-26% and the Hierarchical Attention Network (HAN) by around 4%-12% on both datasets, with around 10%-30% reduction of training time.