Dan Roth

New methods for improving table retrieval enhance accuracy in answering complex queries.

A new benchmark reveals gaps in visual understanding of large language models.

A new framework for evaluating large language models with human insight.

A new framework enhances reasoning in language models through visual sketches.

A new method combines text-based and SQL reasoning for improved table question answering.

This article reviews the reliability of models in analyzing visual data.

Research enhances language models' ability to process time-related information in tables.

This study improves classifier performance through targeted strategies in Multi-Instance Partial Label Learning.

A study on AI's understanding of tables that combine text and images.

MAPWise dataset challenges models on map-based questions and evaluates their reasoning skills.

Context improves the quality and relevance of language model answers.

NAVCON helps machines understand navigation instructions through language and visual cues.