Papers for

biomedical data teams

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

LatentVerse aids understanding of multimodal biomedical data embeddings

LatentVerse: A Framework for Understanding Shared and Modality-Specific Information in Multimodal Latent Representations

Abstract: Latent embeddings have become a central data abstraction in modern machine learning, especially in biomedicine, where foundation models are increasingly used to encode multimodal data like clinical text, medical images, omics, and physiological signals. However, the utility and value of these representations depends on understanding their quality, structure, and the information they encode. Existing analysis workflows for evaluating representations remain fragmented across custom scripts, isolated metrics, and most importantly lack multimodal analysis, limiting accessibility and reproducibility. We present LatentVerse, a representation analysis resource that combines a web-based visual analytics platform for accessible, report-driven exploration with a command-line interface for scalable technical workflows. LatentVerse unifies diagnostics for various representation quality metrics and extends to multimodal settings by decomposing embeddings into shared and modality-specific components. We evaluate LatentVerse through controlled unimodal and multimodal simulations, discovery-oriented analyses on real biomedical embeddings, and a user study across diverse use cases. By supporting thorough and interpretable evaluation of latent spaces, LatentVerse makes foundation model representations more understandable in biomedical and data science applications.

Fri 11 SeptMachine LearningHuman-Computer Interaction
The gist
Machine learning models often turn complex biomedical data into simpler codes called latent embeddings, which mix information from different sources like images and text. The value of these codes depends on how well we understand what information they hold and how it is organized. The authors developed LatentVerse, a tool that helps people explore and analyze these embeddings, especially by separating shared from source-specific information. This makes it easier for users to interpret and trust the results of AI models handling multiple types of biomedical data.
Open 2609.12364v1

Biomedical knowledge graphs help language models generate better hypotheses

HypoKG: Evidence-Disciplined Biomedical Hypothesis Generation Beyond Endpoint Knowledge

Abstract: Large language models (LLMs) can generate biomedical hypotheses, but it remains unclear whether they truly reason from scientific evidence or simply produce convincing-sounding ideas. To study this, we combine three major biological databases: the Kyoto Encyclopedia of Genes and Genomes (KEGG), Rhea, and UniProt, into a unified biochemical knowledge graph and construct a benchmark of 550 paths connecting enzyme sources to rare disease endpoints, yielding 13,200 hypotheses from six LLMs under four conditions varying the biological information each model receives: source enzyme only, full biological path, or source and disease endpoint only. Hypotheses are scored using an expert-derived five-criterion rubric on a 1-5 scale per criterion. We find that models given both the source and disease endpoint often produce the highest-scoring hypotheses, showing that LLMs can generate compelling ideas from minimal information. However, these hypotheses are less grounded in the evidence. In contrast, models given the full biological path generate hypotheses more consistent with known mechanistic relationships. We call this evidence-disciplined reasoning. To confirm this effect, we shuffled intermediate path steps while keeping endpoints fixed. Evidence grounding dropped significantly (delta = -0.793, p < 0.001), confirming models genuinely used path structure during reasoning. Our findings show that knowledge graphs support hypothesis generation in two ways: they identify biological endpoint pairs absent from the literature, and their mechanistic paths guide how LLMs reason between them.

Thu 10 SeptComputation and LanguageArtificial Intelligence
The gist
It can be hard to tell if large language models make good guesses about biology by understanding real evidence or just sounding smart. The authors connected several biology databases into one big network to test how language models generate ideas about diseases and enzymes. They found that when models see the whole biological path, they make more evidence-based guesses, while partial information leads to less grounded but still interesting ideas. This shows that detailed scientific knowledge helps models reason more accurately.
Open 2609.12260v1

Large language models detect and reduce false claims by domain training

Domain-Specific Hallucination Detection in Large Language Models

Abstract: Large language models generate fluent text that can contain unfaithful claims -- a phenomenon known as hallucination. We present a multi-signal detection pipeline combining fine-tuned DeBERTa-v3 classification, Monte Carlo (MC) Dropout uncertainty quantification, and temperature-scaled calibration for response-level hallucination detection. Evaluated on the HaluEval benchmark, our pipeline achieves F1=0.915 and AUROC=0.977 on general-domain tasks, with per-task F1 scores of 0.97 (QA), 0.96 (Summarization), and 0.82 (Dialogue). MC Dropout inference further improves accuracy to 93.2%. A context ablation study confirms the model performs genuine entailment reasoning rather than exploiting surface patterns, with summarization F1 dropping 24% when knowledge context is removed. Learning curve analysis reveals that 25% of training data captures 77% of full-data performance. Beyond detection, we apply Direct Preference Optimization (DPO) to a Qwen2.5-0.5B generator, reducing its hallucination rate from 85.5% to 37.7% (55.9% relative reduction) as measured by our detector. Cross-domain evaluation on the SciFact biomedical benchmark shows that general-domain training transfers poorly (F1=0.52), motivating domain-specific fine-tuning. PubMedBERT fine-tuned on SciFact achieves F1=0.63 and AUROC=0.81, demonstrating that domain-matched pre-training is the strongest adaptation strategy. Code and models are available at https://github.com/varunteja99/hallucination-detection-nlp

Thu 10 SeptComputation and LanguageArtificial IntelligenceMachine Learning
The gist
Large language models sometimes make up information, which is called hallucination. The authors created a system that uses multiple techniques, including a special classifier and uncertainty checks, to spot when these models are likely hallucinating. They tested this system on general and medical topics and found it works best when trained specifically for each area. They also used their detector to help improve another language model, lowering how often it made false claims. This shows the importance of tailoring detection tools to different subject areas.
Open 2609.11878v1

Biomedical retrieval struggles with complex ontology based constraints

OntologyBench: Can Dense Retrieval Satisfy Structured Biomedical Constraints?

Abstract: We introduce OntologyBench, a tiered biomedical retrieval benchmark comprising 471,854 training and 125,744 evaluation query-document relevance pairs across concept grounding, relational retrieval, and compositional phenotype-based retrieval. Although these tasks can be tractable using ontology-aware reference methods, across task tiers, embedding performance is generally lower on relational and compositional tasks than on concept-grounding tasks. Fine-tuning on ontology-derived supervision improves performance on several relational and compositional tasks, whereas the evaluated reranking and LLM-based candidate-scoring methods provide little or no end-to-end improvement. Errors frequently reflect diseases matching only subsets of the phenotype evidence. These findings indicate that the evaluated embedding and reranking configurations do not reliably recover the compatibility encoded by the selected ontology relations and phenotype combinations and motivate retrieval systems that better integrate learned representations with structured biomedical knowledge.

Tue 8 SeptArtificial Intelligence
The gist
Finding relevant medical documents using computerized methods often relies on understanding complex biomedical concepts and their relationships. The paper introduces OntologyBench, a large benchmark to test how well current search methods handle these complex queries. The authors find that simple concept matching works better than finding relations or combinations of medical traits. They also show that fine-tuning improves some results, but other advanced methods do not reliably capture the required detailed biomedical relationships. This suggests that better integration of medical knowledge and learned data representations is needed for improved search.
Open 2609.08174v1