Update README.md

README.md

@@ -9,23 +9,40 @@ pinned: false
 
 # Oxford Reasoning with Machine Learning Lab
 
-We conduct research on **AI evaluation, safety, and human–AI interaction** to advance understanding of how large language models reason, solve complex problems, and collaborate with humans.
+[![Website](https://img.shields.io/badge/⚡️_Website-oxrml.com-002147?style=flat-square)](https://oxrml.com)
+[![Oxford](https://img.shields.io/badge/🎓_University-Oxford-002147?style=flat-square)](https://www.ox.ac.uk)
+[![Focus](https://img.shields.io/badge/🔬_Focus-AI_Evaluation_·_Safety_·_Human--AI-c8962e?style=flat-square)](#)
 
-Our work combines **theoretical rigour** with **empirical investigation** to study how large language models process information, perform tasks, and behave in real-world settings.
+
+> *Combining theoretical rigour with empirical investigation to understand how AI models reason, solve complex problems, and collaborate with humans.*
+
+
+---
+
+## What We Do
+
+We study **how machine learning models process information**, perform tasks, and behave in real-world settings — from the maths of evaluation to the messy reality of human use.
+
+---
 
 ## Research Areas
 
-### Benchmarks and Evaluation
-We study the science of LLM evaluation, using **systematic reviews**, **benchmark analysis**, and **statistical modelling** to examine the validity of existing evaluation practices. We develop new benchmarks and evaluation frameworks to test the limits of LLM reasoning, especially in **adversarial**, **interactive**, and **low-resource language** settings.
+### 📐 Benchmarks & Evaluation
+We study the science of LLM evaluation using systematic reviews, benchmark analysis, and statistical modelling. We develop new benchmarks to test LLM reasoning limits, especially in **adversarial**, **interactive**, and **low-resource language** settings.
+
+### 🔬 Agentic AI for Science
+We build agentic systems that automate and augment key stages of the scientific process: literature discovery, evidence synthesis, hypothesis generation, and decision support. Our agents are **reliable**, **transparent**, and grounded in domain expertise.
 
-### Agentic AI for Science
-We build agentic AI systems that automate and augment key stages of the scientific process, including **literature discovery**, **evidence synthesis**, **hypothesis generation**, and **decision support**. A central focus is developing agents that are **reliable**, **transparent**, and **grounded in domain expertise** for real-world scientific and policy applications.
+### 🛡️ AI Safety
+From bias and toxicity to misalignment in agentic systems: we investigate the harms advanced AI may pose to individuals and society, alongside technical mitigation methods and **AI governance** research.
+
+### 🤝 Human–AI Interaction
+Large-scale empirical studies of how people use and respond to AI systems in **real-world decision-making** contexts.
+
+---
 
-### AI Safety
-We investigate the risks that advanced AI systems may pose to individuals and society. Our work spans the spectrum of harms, from **bias and toxicity in language models** to **misalignment in agentic systems**, alongside technical methods for mitigation and research on AI governance.
+## 🏷️ Topics
 
-### Human–AI Interaction
-We conduct large-scale empirical studies of how people use and respond to AI systems in decision-making contexts.
+`llm-evaluation` `benchmarking` `ai-safety` `agentic-ai` `human-ai-interaction` `reasoning` `nlp` `alignment` `bias` `governance` `low-resource-nlp` `scientific-discovery`
 
-## Links
-- Website: [oxrml.com](https://oxrml.com/)
+---