EXPLORE-Bench: Egocentric Scene Prediction with Long-Horizon Reasoning

TL;DR

This paper introduces EXPLORE-Bench, a new benchmark for evaluating multimodal large language models on their ability to perform long-horizon egocentric scene prediction, revealing significant gaps compared to human reasoning.

Contribution

The paper presents a novel benchmark, EXPLORE-Bench, for systematic evaluation of long-term egocentric scene prediction by multimodal models, with detailed annotations and diverse real-world scenarios.

Findings

Models lag behind humans in long-horizon egocentric reasoning.

Decomposing action sequences can improve model performance.

Stepwise reasoning incurs computational overhead.

Abstract

Multimodal large language models (MLLMs) are increasingly considered as a foundation for embodied agents, yet it remains unclear whether they can reliably reason about the long-term physical consequences of actions from an egocentric viewpoint. We study this gap through a new task, Egocentric Scene Prediction with LOng-horizon REasoning: given an initial-scene image and a sequence of atomic action descriptions, a model is asked to predict the final scene after all actions are executed. To enable systematic evaluation, we introduce EXPLORE-Bench, a benchmark curated from real first-person videos spanning diverse scenarios. Each instance pairs long action sequences with structured final-scene annotations, including object categories, visual attributes, and inter-object relations, which supports fine-grained, quantitative assessment. Experiments on a range of proprietary and open-source MLLMs reveal a significant performance gap to humans, indicating that long-horizon egocentric reasoning remains a major challenge. We further analyze test-time scaling via stepwise reasoning and show that decomposing long action sequences can improve performance to some extent, while incurring non-trivial computational overhead. Overall, EXPLORE-Bench provides a principled testbed for measuring and advancing long-horizon reasoning for egocentric embodied perception.