Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2509.19524 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911173068193792 |
|---|---|
| author | ElMallah, Ramy Chhajer, Krish Lee, Chi-Guhn |
| author_facet | ElMallah, Ramy Chhajer, Krish Lee, Chi-Guhn |
| contents | Robot learning papers typically report a single binary success rate (SR), which obscures where a policy succeeds or fails along a multi-step manipulation task. We argue that subgoal-level reporting should become routine: for each trajectory, a vector of per-subgoal SRs that makes partial competence visible (e.g., grasp vs. pour). We propose a blueprint for StepEval, a cost-aware plug-in evaluation framework that utilizes vision-language models (VLMs) as automated judges of subgoal outcomes from recorded images or videos. Rather than proposing new benchmarks or APIs, our contribution is to outline design principles for a scalable, community-driven open-source project. In StepEval, the primary artifact for policy evaluation is the per-subgoal SR vector; however, other quantities (e.g., latency or cost estimates) are also considered for framework-optimization diagnostics to help the community tune evaluation efficiency and accuracy when ground-truth subgoal success labels are available. We discuss how such a framework can remain model-agnostic, support single- or multi-view inputs, and be lightweight enough to adopt across labs. The intended contribution is a shared direction: a minimal, extensible seed that invites open-source contributions, so that scoring the steps, not just the final goal, becomes a standard and reproducible practice. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_19524 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Score the Steps, Not Just the Goal: VLM-Based Subgoal Evaluation for Robotic Manipulation ElMallah, Ramy Chhajer, Krish Lee, Chi-Guhn Artificial Intelligence Robotics Robot learning papers typically report a single binary success rate (SR), which obscures where a policy succeeds or fails along a multi-step manipulation task. We argue that subgoal-level reporting should become routine: for each trajectory, a vector of per-subgoal SRs that makes partial competence visible (e.g., grasp vs. pour). We propose a blueprint for StepEval, a cost-aware plug-in evaluation framework that utilizes vision-language models (VLMs) as automated judges of subgoal outcomes from recorded images or videos. Rather than proposing new benchmarks or APIs, our contribution is to outline design principles for a scalable, community-driven open-source project. In StepEval, the primary artifact for policy evaluation is the per-subgoal SR vector; however, other quantities (e.g., latency or cost estimates) are also considered for framework-optimization diagnostics to help the community tune evaluation efficiency and accuracy when ground-truth subgoal success labels are available. We discuss how such a framework can remain model-agnostic, support single- or multi-view inputs, and be lightweight enough to adopt across labs. The intended contribution is a shared direction: a minimal, extensible seed that invites open-source contributions, so that scoring the steps, not just the final goal, becomes a standard and reproducible practice. |
| title | Score the Steps, Not Just the Goal: VLM-Based Subgoal Evaluation for Robotic Manipulation |
| topic | Artificial Intelligence Robotics |
| url | https://arxiv.org/abs/2509.19524 |