MultivationBench - Example F1: leaderboard
MultivationBench (HKUST, Findings of EMNLP 2026) scores multi-label motivation predictions for character behaviors in story-driven visual narratives with partial credit. The score is example-based F1 in percent, on the common subset of tasks every model answered; higher is better.
Metric: Example-based F1 (%, partial credit per question) of multi-label motivation predictions for visually grounded character behaviors in sequential visual narratives, reasoning over the accumulated story images and text (Maslow 8-level needs and Reiss 16 basic desires, definition and practical-motivation tasks), zero-shot at temperature 0; multimodal input, common subset of 14,180 tasks from 1,000 stories that every model answered; higher is better. Source: arxiv.org. Saturation forecast: Around May 2028. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Grok 4.1 Fast | 55.32 | #430 |
| 2 | Gemini 3 Flash | 54.22 | #164 |
| 3 | O4 Mini | 53.62 | #353 |
| 4 | Llama 4 Maverick | 49.81 | #1099 |
| 5 | Llama 4 Scout | 49.6 | #1776 |
| 6 | Phi-4 Multimodal Instruct | 43.03 | #3882 |
| 7 | Nemotron Nano 12B v2 VL | 38.16 | #2231 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=multivationbench-example-f1 · How It Works · Data refreshed daily, snapshot 2026-11-13.