Causal-Plan-Bench - Robustness: leaderboard

Metric: Robustness score: mean rubric score (0-100): a blind GPT-5.6-sol judge scores each open-ended answer against a human-verified reference from 0 to 1, times 100, over the plan diagnosis and repair, counterfactual outcome and failure recovery tasks; 100 held-out expert-reviewed items per task drawn from egocentric and robot video (EPIC-KITCHENS, Ego4D, HoloAssist and others), the same instruction, visual evidence and answer format for every model, temperature 0.2, mean of three regenerated runs; higher is better. Source: arxiv.org. Saturation forecast: Around 2035. 14 models tracked.

Top models

#ModelScore
1GPT-644.17
2Gemini 3.1 Pro (Preview)38.9
3GPT-5.6 Sol38.57
4Seed 2.1 Pro37.6
5Kimi K2.535.93
6Qwen 3 VL 8B Instruct33.13
7InternVL3.5-8B32.17
8GPT-4o32.07

Interactive version: theaggregate.ai/benchmark?slug=causal-plan-bench-robustness · How It Works · Data refreshed daily, snapshot 2026-09-29.