Hanyang University

2 related articles

Subscribe to updates

researchsquare.com Oct 23, 2026
Nanoparticle safety stops being a material-only question: a host-response axis lifts biosafety AUC from 0.66 to 0.90
Nanoparticle neurological safety is hard to predict from material properties alone, because exposure conditions, disease context and host responses reshape outcomes. Across 893 experimental contexts from 599 neurological studies, pro-inflammatory, anti-inflammatory/resolution and apoptosis-associated responses were harmonized into a Conserved Response Axis and added layer by layer to material descriptors. With a Random Forest model, biosafety ROC-AUC rose from 0.66 with material properties alone to 0.88 with exposure and biological context and 0.90 with cellular responses; recovery rose from 0.64 to 0.77 and 0.78.
Nanoparticle neurological safety is hard to predict from material properties alone, because exposure conditions, disease context and host responses reshape outcomes. Across 893 experimental contexts from 599 neurological studies, pro-inflammatory, anti-inflammatory/resolution and apoptosis-associated responses were harmonized into a Conserved Response Axis and added layer by layer to material descriptors. With a Random Forest model, biosafety ROC-AUC rose from 0.66 with material properties alone to 0.88 with exposure and biological context and 0.90 with cellular responses; recovery rose from 0.64 to 0.77 and 0.78.
Sejong University Ulsan National Institute of Science and Technology B.S. Abdur Rahman Crescent Institute of Science & Technology Kyung Hee University Incheon National University Chalmers University of Technology Hanyang University Technical University of Denmark
Sumit Kumar（First author, Corresponding author）Seungah Lee（Corresponding author）Haribalan Perumalsamy（Corresponding author）
Conserved Response Axis Prediction of Biosafety and Recovery Hierarchical Machine-Learning Framework Random Forest SHAP
arXiv Oct 1, 2026
EvoDuet lets LLMs retrieve the web by knowledge gap during evolutionary search, raising OpenEvolve's normalized discovery gain from 74.1% to 78.0% (GPT-5.6-Luna) and from 61.3% to 82.3% (Gemini-3.8-Flash) across 21 optimization tasks and surpassing prior best scores on eight
The work introduces EvoDuet, a bi-level method that co-evolves solutions and web search queries with fixed model parameters: at each iteration a knowledge-gap-based retrieval gate decides whether to retrieve new documents, reuse stored ones, or proceed without them, an inner loop refines queries and ranks documents by the solution scores they are predicted to yield, and an outer loop generates candidates in parallel and records evaluated outcomes for later searches; across 21 optimization tasks with one candidate per iteration it raises OpenEvolve's normalized discovery gain from 74.1% to 78.0% with GPT-5.6-Luna and from 61.3% to 82.3% with Gemini-3.8-Flash, Qwen3.
The work introduces EvoDuet, a bi-level method that co-evolves solutions and web search queries with fixed model parameters: at each iteration a knowledge-gap-based retrieval gate decides whether to retrieve new documents, reuse stored ones, or proceed without them, an inner loop refines queries and ranks documents by the solution scores they are predicted to yield, and an outer loop generates candidates in parallel and records evaluated outcomes for later searches; across 21 optimization tasks with one candidate per iteration it raises OpenEvolve's normalized discovery gain from 74.1% to 78.0% with GPT-5.6-Luna and from 61.3% to 82.3% with Gemini-3.8-Flash, Qwen3.
University of Minnesota Hanyang University Seoul National University Korea Advanced Institute of Science and Technology
Young-Jun Lee（First author）
EvoDuet Knowledge-Gap-Based Retrieval Gating Hypothetical Evidence Scoring Evolutionary Search Scientific Discovery

You’re all caught up

Nanoparticle safety stops being a material-only question: a host-response axis lifts biosafety AUC from 0.66 to 0.90

EvoDuet lets LLMs retrieve the web by knowledge gap during evolutionary search, raising OpenEvolve's normalized discovery gain from 74.1% to 78.0% (GPT-5.6-Luna) and from 61.3% to 82.3% (Gemini-3.8-Flash) across 21 optimization tasks and surpassing prior best scores on eight