Author

V. R. Parupudi

h-index 1 View on Semantic Scholar

We have 1 of 4 papers

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Preprint Jul 2026

There Is No Neutral Harness: Modern LLM Leaderboards Are Manufactured by Config-Fragile Items

This work treats the evaluation harness of large language models (LLMs) as an independent variable and resolve its effect to single items, and introduces the fragility grid, which is a check a leaderboard can run before it reports an order.

V. R. Parupudi · 0 citations