OpenAlex ·

FailBench: How Reliable are VLMs at Judging Robot Task Success?

Zaruhi Navasardyan, Tatul Danielyan and Hrant Davtyan

Published
Source
OpenAlex
Published in
arXiv (Cornell University)
DOI
10.48550/arxiv.2609.03611
Original
Read it at the publisher
Topics
Computer VisionSystems & Control

Abstract

Vision-Language Models (VLMs) are increasingly used to evaluate robot manipulation outcomes, but existing benchmarks offer limited evidence of cross-domain generalization. We introduce FailBench, a benchmark for robot failure detection comprising 2,197 manipulation attempts across 14 public sources (12 real-world, 2 simulated). In FailBench, 75% of failures occur naturally, and six real-world sources come from non-failure-detection datasets. Evaluating 13 VLM-based detectors, we find the best model achieves only 0.77 mean balanced accuracy. Notably, models fine-tuned for failure detection consistently underperform general-purpose VLMs and their own pretrained baselines. Performance depends heavily on required visual evidence: models approach saturation when outcomes depend on observable object motion, but degrade to near-chance (<0.60 balanced accuracy) on contact-intensive assembly tasks. Error analysis reveals a systematic bias toward predicting success under ambiguous evidence, which persists even with increased reasoning effort. Finally, we show that input-level intervention--spatially localizing and cropping outcome-relevant regions--improves the top detector by 2.4 percentage points without extra training.

The authors’ abstract, as arXiv lists it. arXiv’s listings, abstracts included, are free for anyone to reuse.

Read it on ReadSearch

Each opens after you create an account, and brings you back to this paper.

Already a reader? Sign in

All new papers in Computer Vision

Get the few papers worth your time in Computer Vision every morning.

ReadSearch reads the new papers in your field each day and brings you the few worth reading, each with its key finding in a sentence.

Papers belong to their authors and publishers; every page here links to the original. All research topics