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

• URL: http://arxiv.org/abs/2609.03611v1
• Date: Thu, 03 Sep 2026 09:58:55 GMT
• Title: FailBench: How Reliable are VLMs at Judging Robot Task Success?
• Abstract summary: FailBench is a benchmark for robot failure detection comprising 2,197 manipulation attempts across 14 public sources.<n>In FailBench, 75% of failures occur naturally, and six real-world sources come from non-failure-detection datasets.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• 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.

Related papers

• Learning from Failure: Inference-Time Self-Improvement for Computer-Use Agents [80.03824805228719]
  A major challenge in developing computer-use agents is collecting large-scale, high-quality trajectories.<n>We propose a data-centric paradigm that turns failed trajectories into agent improvements.<n>We validate this approach with the state-of-the-art OpenCUA-72B model on the OSWorld benchmark.
  arXiv  Detail & Related papers  (2026-06-30T07:44:37Z)
• FATE-VLA:Failue-aware test generation for vision-language-action models [10.716019832991046]
  Vision-Language-Action models are increasingly used as generalist robot policies.<n>In high-dimensional embodied spaces, failures are sparse and clustered, so static benchmarking can underestimate risks.<n>We propose a failure-aware test-generation approach that combines diversity-driven exploration with surrogate models learned from observed executions.
  arXiv  Detail & Related papers  (2026-06-01T14:27:13Z)
• Rewarding the Scientific Process: Process-Level Reward Modeling for Agentic Data Analysis [68.28714988482703]
  Process Reward Models (PRMs) have achieved remarkable success in augmenting the reasoning capabilities of Large Language Models (LLMs)<n>We present a study revealing that general-domain PRMs struggle to supervise data analysis agents.<n>We introduce DataPRM, a novel environment-aware generative process reward model.
  arXiv  Detail & Related papers  (2026-04-27T09:00:30Z)
• Failure Identification in Imitation Learning Via Statistical and Semantic Filtering [2.2515832864485024]
  Imitation learning (IL) policies in robotics deliver strong performance in controlled settings but remain brittle in real-world deployments.<n>We introduce FIDeL (Failure Identification in Demonstration Learning), a policy-independent failure detection module.<n>We also introduce BotFails, a multimodal dataset of real-world tasks for failure detection in robotics.
  arXiv  Detail & Related papers  (2026-04-15T12:27:32Z)
• From Passive Observer to Active Critic: Reinforcement Learning Elicits Process Reasoning for Robotic Manipulation [32.41847293364159]
  PRIMO R1 is a 7B framework that transforms video MLLMs into active "Critics"<n>We leverage outcome-based Reinforcement Learning to incentivize explicit Chain-of-Thought generation for progress estimation.<n>We establish state-of-the-art performance on RoboFail benchmark with 67.0% accuracy, surpassing closed-source models like OpenAI o1 by 6.2%.
  arXiv  Detail & Related papers  (2026-03-16T17:53:28Z)
• Outcome Accuracy is Not Enough: Aligning the Reasoning Process of Reward Models [108.26461635308796]
  We introduce Rationale Consistency, a fine-grained metric that quantifies the alignment between the model's reasoning process and human judgment.<n>Our evaluation of frontier models reveals that rationale consistency effectively discriminates among state-of-the-art models.<n>We introduce a hybrid signal that combines rationale consistency with outcome accuracy for GenRM training.
  arXiv  Detail & Related papers  (2026-02-04T15:24:52Z)
• Reshaping Action Error Distributions for Reliable Vision-Language-Action Models [69.38615670891038]
  In robotic manipulation, vision-language-action (VLA) models have emerged as a promising paradigm for learning generalizable and scalable robot policies.<n>We focus on continuous-action VLA models and move beyond conventional MSE-based regression by reshaping action error distributions during training.<n>We evaluate our approaches across standard, few-shot, and noisy settings on multiple representative VLA architectures.
  arXiv  Detail & Related papers  (2026-02-04T05:37:09Z)
• Guardian: Detecting Robotic Planning and Execution Errors with Vision-Language Models [53.20969621498248]
  We propose an automatic robot failure synthesis approach that procedurally perturbs successful trajectories to generate diverse planning and execution failures.<n>We construct three new failure detection benchmarks: RLBench-Fail, BridgeDataV2-Fail, and UR5-Fail.<n>We then train Guardian, a VLM with multi-view images for detailed failure reasoning and detection.
  arXiv  Detail & Related papers  (2025-12-01T17:57:27Z)
• Leveraging Unlabeled Data to Predict Out-of-Distribution Performance [63.740181251997306]
  Real-world machine learning deployments are characterized by mismatches between the source (training) and target (test) distributions. In this work, we investigate methods for predicting the target domain accuracy using only labeled source data and unlabeled target data. We propose Average Thresholded Confidence (ATC), a practical method that learns a threshold on the model's confidence, predicting accuracy as the fraction of unlabeled examples.
  arXiv  Detail & Related papers  (2022-01-11T23:01:12Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.