Chitta Baral

This study focuses on enhancing spatial accuracy in text-to-image generation.

TPO offers a new method to align language models with human preferences efficiently.

Research seeks to improve how LLMs handle misleading information.

This article examines how negation affects large language models and their accuracy.

New dataset assesses LLMs' ability for complex logical reasoning tasks.

A novel test for evaluating reasoning about timing without relying on prior knowledge.

An analysis of LLM performance on grid puzzles to assess reasoning abilities.

A new framework enhances how models understand spatial relationships in images.