Affordance Agent Harness: Verification-Gated Skill Orchestration

TL;DR

This paper introduces Affordance Agent Harness, a closed-loop system that adaptively orchestrates multiple skills for improved affordance grounding in complex scenes, balancing accuracy and inference cost.

Contribution

It presents a novel unified framework with evidence management, episodic memory, and adaptive skill routing, outperforming fixed pipelines in accuracy and efficiency.

Findings

Achieves better accuracy-cost trade-offs than fixed pipelines.

Reduces average skill calls and latency in affordance grounding.

Improves robustness in challenging, occluded, and ambiguous scenes.

Abstract

Affordance grounding requires identifying where and how an agent should interact in open-world scenes, where actionable regions are often small, occluded, reflective, and visually ambiguous. Recent systems therefore combine multiple skills (e.g., detection, segmentation, interaction-imagination), yet most orchestrate them with fixed pipelines that are poorly matched to per-instance difficulty, offer limited targeted recovery from intermediate errors, and fail to reuse experience from recurring objects. These failures expose a systems problem: test-time grounding must acquire the right evidence, decide whether that evidence is reliable enough to commit, and do so under bounded inference cost without access to labels. We propose Affordance Agent Harness, a closed-loop runtime that unifies heterogeneous skills with an evidence store and cost control, retrieves episodic memories to provide priors for recurring categories, and employs a Router to adaptively select and parameterize skills. An affordance-specific Verifier then gates commitments using self-consistency, cross-scale stability, and evidence sufficiency, triggering targeted retries before a final judge fuses accumulated evidence and trajectories into the prediction. Experiments on multiple affordance benchmarks and difficulty-controlled subsets show a stronger accuracy-cost Pareto frontier than fixed-pipeline baselines, improving grounding quality while reducing average skill calls and latency. Project page: https://tenplusgood.github.io/a-harness-page/.