AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

CLINIC : Evaluating Multilingual Trustworthiness in Language Models for Healthcare

Akash Ghosh, Srivarshinee Sridhar, Raghav Kaushik Ravi, Muhsin Muhsin, Sriparna Saha, Chirag Agarwal

ICML 2026 · Large language models › Language, safety and interpretability · approach: Multilingual LLM data and culture · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
move from high- to low-resource languagePhi-4mini · Honestyhonesty rate: 83.50 -> 24.50Hurt
move from high- to low-resource languageGPT-4o-mini · Honestyhonesty rate: 78.38 -> 68.50Hurt
move from high- to low-resource languageGemini-2.5-Pro · Honestyhonesty rate: 95.20 -> 93.00Hurt
move from high- to low-resource languageMMedLLama · Consistencysimilarity: 0.657 -> 0.573Hurt
move from high- to low-resource languageDSeek-R1 · Consistencysimilarity: 0.749 -> 0.680Hurt
move from high- to low-resource languageGPT-4o-mini · Consistencysimilarity: 0.781 -> 0.743Hurt
use medical fine-tuned model instead of general modelOpenBioLLM-8B vs LLaMA-3.2-3B · Honestyhonesty rate: 75.50 -> 40.75Hurt
move from high- to mid-resource languageGemini-2.5-Pro · OOD drug RtARtA: 90.87 -> 97.33Improved
move from high- to low-resource languageQwen-2-1.5B · OOD drug RtARtA: 62.50 -> 41.67Hurt
move from high- to low-resource languageQwen3-32B · Sycophancysimilarity score: 0.054 -> 0.018Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy