Self Driving Datasets: From 20 Million Papers to Nuanced Biomedical Knowledge at Scale

• URL: http://arxiv.org/abs/2605.07022v1
• Date: Thu, 07 May 2026 23:08:18 GMT
• Title: Self Driving Datasets: From 20 Million Papers to Nuanced Biomedical Knowledge at Scale
• Abstract summary: We show that PubMed can be autonomously and cost-effectively turned into larger, more nuanced, and more accurate structured datasets.<n>We present three contributions: (1) an entity-tagging pipeline, grounded in nine6.3 biomedical repositories; (2) hybrid retrieval supporting entity-targeted corpus repositories; and (3) Starling, a deep research system given only a natural-language description.
• Score: 34.468123235616524
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Manually curated biomedical repositories -- spanning bioactivity, genomics, and chemistry -- are expensive to maintain, lag behind primary literature, and discard experimental context, obscuring nuances needed to assess data correctness and coverage. We show that PubMed itself can be autonomously and cost-effectively turned into structured datasets that are larger, more nuanced, and more accurate than the curated databases they replace. We present three coupled contributions: (1) an LLM-based entity-tagging pipeline, grounded in nine biomedical ontologies, that tags 4.5B entities across 19 categories in a 22.5M-paper, 2.5T-token PubMed corpus; (2) hybrid sparse-dense retrieval supporting entity-filtered semantic queries over the tagged corpus; and (3) Starling, a multi-agent deep research system that, given only a natural-language task description, designs precision- and recall-targeted retrieval filters, induces an extraction schema, and emits structured records with nuance-rich fields and supporting passages. Across six tasks -- blood-brain barrier permeability, oral bioavailability, acute toxicity (LD50), gene-disease associations, protein subcellular localization, and chemical reactions -- Starling produces ~6.3M records (91K-3M per task); several are, to our knowledge, the largest public datasets for their property. Frontier-model rejection of our extractions is 0.6-7.7% across tasks, far below error rates we measure on widely used curated counterparts (e.g., 16.5% on BBB_Martins, 7.3% on Bioavailability_Ma). Beyond scale and accuracy, the supporting passages carry nuance tabular databases discard -- e.g., oral bioavailability may depend on fed vs. fasted state. Together, the corpus, retrieval, and agent establish a foundation for AI-driven therapeutic design. Code and datasets: https://github.com/starling-labs/starling.

Related papers

• bioMoR: Biology-Guided Mixture-of-Recursions for Effective Genomic Learning [11.154086943903694]
  Mixture-of-Recursions (MoR) improves efficiency through adaptive token-choice or expert-choice routing.<n>We propose bioMoR, which is the first framework to apply MoR to gene-level and pathway-level learning.
  arXiv  Detail & Related papers  (2026-08-07T02:44:11Z)
• BioMiner: A Multi-modal System for Automated Mining of Protein-Ligand Bioactivity Data from Literature [53.894504720119805]
  We introduce BioMiner, a multi-modal extraction framework for bioactivity data extraction.<n>Within BioMiner, bioactivity semantics are inferred through direct reasoning, while chemical structures are resolved via a chemical-structure-grounded visual semantic reasoning paradigm.<n>For rigorous evaluation and method development, we establish a benchmark comprising 16,457 bioactivity entries curated from 500 publications.
  arXiv  Detail & Related papers  (2026-04-23T10:11:56Z)
• Enhancing Biomedical Named Entity Recognition using GLiNER-BioMed with Targeted Dictionary-Based Post-processing for BioASQ 2025 task 6 [0.0]
  This study evaluates the GLiNER-BioMed model on a BioASQ dataset.<n>We introduce a targeted dictionary-based post-processing strategy to address common misclassifications.<n>This work highlights the potential of dictionary-based refinement for pre-trained BioNER models but underscores the critical challenge of overfitting to development data.
  arXiv  Detail & Related papers  (2025-10-03T18:35:04Z)
• TheBlueScrubs-v1, a comprehensive curated medical dataset derived from the internet [1.4043931310479378]
  TheBlueScrubs-v1 is a curated dataset of over 25 billion medical tokens drawn from a broad-scale internet corpus.<n>Each text is assigned three LLM-based quality scores encompassing medical relevance, precision and factual detail, and safety and ethical standards.<n>This Data Descriptor details the dataset's creation and validation, underscoring its potential utility for medical AI research.
  arXiv  Detail & Related papers  (2025-04-01T22:25:19Z)
• BIOMEDICA: An Open Biomedical Image-Caption Archive, Dataset, and Vision-Language Models Derived from Scientific Literature [73.39593644054865]
  BIOMEDICA is a scalable, open-source framework to extract, annotate, and serialize the entirety of the PubMed Central Open Access subset into an easy-to-use, publicly accessible dataset.<n>Our framework produces a comprehensive archive with over 24 million unique image-text pairs from over 6 million articles.<n> BMCA-CLIP is a suite of CLIP-style models continuously pretrained on the BIOMEDICA dataset via streaming, eliminating the need to download 27 TB of data locally.
  arXiv  Detail & Related papers  (2025-01-13T09:58:03Z)
• SMILE-UHURA Challenge -- Small Vessel Segmentation at Mesoscopic Scale from Ultra-High Resolution 7T Magnetic Resonance Angiograms [60.35639972035727]
  The lack of publicly available annotated datasets has impeded the development of robust, machine learning-driven segmentation algorithms. The SMILE-UHURA challenge addresses the gap in publicly available annotated datasets by providing an annotated dataset of Time-of-Flight angiography acquired with 7T MRI. Dice scores reached up to 0.838 $pm$ 0.066 and 0.716 $pm$ 0.125 on the respective datasets, with an average performance of up to 0.804 $pm$ 0.15.
  arXiv  Detail & Related papers  (2024-11-14T17:06:00Z)
• BiomedRAG: A Retrieval Augmented Large Language Model for Biomedicine [19.861178160437827]
  Large Language Models (LLMs) have swiftly emerged as vital resources for different applications in the biomedical and healthcare domains. textscBiomedRAG attains superior performance across 5 biomedical NLP tasks. textscBiomedRAG outperforms other triple extraction systems with micro-F1 scores of 81.42 and 88.83 on GIT and ChemProt corpora, respectively.
  arXiv  Detail & Related papers  (2024-05-01T12:01:39Z)
• BMRetriever: Tuning Large Language Models as Better Biomedical Text Retrievers [48.21255861863282]
  BMRetriever is a series of dense retrievers for enhancing biomedical retrieval. BMRetriever exhibits strong parameter efficiency, with the 410M variant outperforming baselines up to 11.7 times larger.
  arXiv  Detail & Related papers  (2024-04-29T05:40:08Z)
• Benchmarking large language models for biomedical natural language processing applications and recommendations [22.668383945059762]
  Large Language Models (LLMs) have shown promise in general domains.<n>We compare their zero-shot, few-shot, and fine-tuning performance with traditional fine-tuning of BERT or BART models.<n>We find issues like missing information and hallucinations in LLM outputs.
  arXiv  Detail & Related papers  (2023-05-10T13:40:06Z)
• Discovering Drug-Target Interaction Knowledge from Biomedical Literature [107.98712673387031]
  The Interaction between Drugs and Targets (DTI) in human body plays a crucial role in biomedical science and applications. As millions of papers come out every year in the biomedical domain, automatically discovering DTI knowledge from literature becomes an urgent demand in the industry. We explore the first end-to-end solution for this task by using generative approaches. We regard the DTI triplets as a sequence and use a Transformer-based model to directly generate them without using the detailed annotations of entities and relations.
  arXiv  Detail & Related papers  (2021-09-27T17:00:14Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.