Spaces:

---

vinid

/

kable-leaderboard

like 2

Sleeping

Hi 👋

Epistemic reasoning evaluation is a fascinating and underexplored area. A model's ability to know what it doesn't know directly correlates with its hallucination rate — and I built a framework that measures both sides of this.

LLM Evaluation Framework tracks:

→ 🔍 Hallucination Rate — measures overconfident wrong outputs (anti-epistemic behavior)
→ 🧠 Reasoning Quality — chain-of-thought depth, which supports epistemic transparency
→ 🎯 Accuracy — task completion
→ 💰 Cost per 1K tokens
→ ⚡ Latency

The combination of epistemic reasoning score + hallucination rate + reasoning quality gives a much richer picture of model calibration.

Live demo: https://huggingface.co/spaces/vigneshwar234/llm-eval-demo
GitHub: https://github.com/vignesh2027/LLM-Evaluation-Framework