Mohit Bansal

This method enhances coding accuracy through reusable function learning and refactoring.

A novel method enhances Vision-and-Language Navigation using training data from driving videos.

Addressing ethical concerns through selective memory removal in AI models.

Study reveals challenges and progress in chatbot memory during lengthy dialogues.

A new method boosts video question-answering while retaining prior knowledge.

Introducing a flexible framework for improved image and video generation control.

A new system streamlines video editing through automated descriptions.

A new method improves how machines comprehend long videos.

A new training method enhances how language models express certainty in answers.

This article examines if language models possess beliefs and follow coherence norms.

A new approach enhances accuracy and creativity in language model outputs.

Examining how cultural bias affects AI image understanding.

Exploring how machines can follow human directions in real-world spaces.

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

Study enhances video understanding through motion-focused questions and visual answers.

Research focuses on teaching machines to follow spoken and written navigation instructions.

Learn how new methods improve timing accuracy in video analysis.