Nanyun Peng

A new benchmark improves how we assess LVLMs and their accuracy.

A new method enhances self-training for language agents using reflection models.

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

Introducing a new scale for evaluating emotional depth in storytelling.

LLM-A* merges traditional algorithms with language models for efficient path planning.

A new system enhances accuracy and reliability in text generation from RALMs.

Analyzing the storytelling capabilities of large language models compared to human authors.

New methods improve automated question-answer relationships in writing analysis.

Enhancing egocentric video learning through transformed exocentric data.

ARMADA improves image-text pairing through attribute-focused data creation.

A new model transforms plain texts into fitting song lyrics.

Research shows ways to enhance context awareness in language models for better responses.

A new method helps computers identify objects using fewer images and simple language.

A safety database to respect cultural norms and legal guidelines globally.