Toward Low-Latency Vision-Language Models with Doubly-Correct Predictions in Egocentric Visual Understanding

2026-06-23

AI Summary

Existing pruning methods for vision-language models in human-robot collaboration scenarios struggle to simultaneously preserve prediction accuracy and evidential interpretability, thereby compromising safety and trustworthiness in low-latency applications. This work proposes a novel rationale-based pruning strategy that explicitly aligns model decisions with evidence localization while compressing the model to meet real-time requirements. Framed around the concept of “dual-correct prediction”—defined as both accurate prediction and traceable evidential support—the approach significantly outperforms current pruning techniques on egocentric video understanding tasks. It achieves not only the highest prediction accuracy but also substantial improvements on the dual-correct prediction metric, demonstrating enhanced reliability and transparency in model reasoning.

Abstract

The rapid rise of Vision-Language Models (VLMs) in egocentric visual understanding has made low-latency inference in human-robot collaborative (HRC) tasks increasingly critical. Weight pruning techniques developed for VLMs to shrink model size and computation can be readily applied to satisfy the efficiency demands of on-board processing and real-time interactive robotics. Moreover, safe human-robot interaction demands pruning strategies that preserve doubly-correct predictions; outputs must be both accurate and evidentially grounded to mitigate risks and ensure user trust. In this paper, we present a new study of VLM pruning through the lens of doubly-correct prediction. Our experiments surprisingly show that existing pruning methods often preserve the right evidence localization but undermine correct prediction. To address this, we propose a rationale-informed pruning strategy that better aligns evidence with decisions. Benchmark results on egocentric video datasets demonstrate that our method not only achieves the highest prediction accuracy but also outperforms existing approaches in attaining doubly-correct predictions. We aim to stimulate research on efficient and reliable VLMs, ensuring accuracy-driven advances align with the transparency, auditability, and safety required for responsible human-robot interaction and embodied intelligence.

Problem

• Vision-Language Models
• Low-Latency
• Doubly-Correct Predictions
• Egocentric Visual Understanding
• Human-Robot Collaboration

Innovation

• doubly-correct prediction
• rationale-informed pruning
• vision-language models
• egocentric visual understanding
• model compression