← Search

Arjun Patrawala

2 accepted papers

2025

Uncovering Gaps in How Humans and LLMs Interpret Subjective Language

ICLR 2025spotlight

Humans often rely on subjective natural language to direct language models (LLMs); for example, users might instruct the LLM to write an *enthusiastic* blogpost, while developers might train models to be *helpful* and *harmless* using LLM-based edits. The LLM’s *operational semantics* of such subjec…

Cited by 0SourcePDFScholar