Author

Shikhar Shiromani

h-index 3 View on Semantic Scholar

We have 2 of 15 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 Aug 2026

A False Average: Pooled CoT-Monitor Accuracy Conceals a Reasoning-Dependent Fragility

This work exposes the fragility of chain-of-thought monitors with a controlled attack: it rewrite only the agent's reasoning to read as good-faith engineering, leaving every command and output byte-identical, so the exploit is unchanged.

Shikhar Shiromani, Leo Richter · 0 citations

Preprint Aug 2026

A False Average: Chain-of-Thought Monitors Collapse Where They Are the Only Defense

This work shows that chain-of-thought monitoring is meant to catch the reward hacks that look clean in the actions and betray themselves only in the reasoning, and it is shown that this is exactly where an adversary who controls the reasoning can de-feat it.

Shikhar Shiromani, Leo Richter · 0 citations