ESCA: Contextualizing Embodied Agents via Scene-Graph Generation

NeurIPSSpotlight2025

Authors: Jiani Huang, Amish Sethi, Matthew Kuo, Mayank Keoliya, Neelay Velingker, JungHo Jung, Ser-Nam Lim, Ziyang Li, Mayur Naik
Venue: NeurIPS 2025
Track: Spotlight

TL;DR

Multi-modal large language models (MLLMs) are making rapid progress toward general-purpose embodied agents…

Opening excerpt from the authors’ abstract. source

Read the paper

OpenReview

Topics

large language model language model agent graph llm

← All NeurIPS 2025 Spotlight papers · Browse the whole archive