HarnessEval-W: Agentifying the Evaluation of Visual Worlds
A benchmark should deliver more than a scalar score: what makes an evaluation trustworthy is the reasoning that justifies the score. This is especially critical for world models, where judging a rollout requires understanding whether physics, causality, and world state evolve correctly. Humans spot such violations naturally, yet no existing benchmark automates this capability: metrics are computed brute-force, leaving no reasoning chain that can be examined or verified. We introduce HarnessEval-W, an agentified evaluation pipeline that brings the harness paradigm from the LLM ecosystem to world model benchmarking. Rather than applying a fixed rubric, HarnessEval-W interprets the context o...
| Authors | Weiliang Chen, Haowen Sun, Jun Gao et al. |
|---|---|
| Primary category | cs.CV |
| Published | 17 August 2026 |
| Last updated at source | 1 September 2026 |
Connected in the knowledge graph
- MirroS-Lab/HarnessEval-WDataset · cited by
Models and datasets referencing this paper
- MirroS-Lab/HarnessEval-WDataset
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.