LetTokenBurnMM-OCEAN: A Multi-Granularity Benchmark for Grounded Personality Reasoning 🔥 News [2026-05] MM-OCEAN was featured as the #2 Paper of the Day on Hugging Face Daily Papers!
$ npm i -g lettokenburn-cli$ ltb bench info mm-ocean
External harness. No Harbor adapter has been generated for this benchmark yet. `ltb bench info` prints the known links.
Runs happen on your machine with your own subscription; only the score summary is pushed. CLI docs
No sourced scores recorded yet.
Every number links to where it was published. Low-confidence rows are muted and need per-cell verification.
MM-OCEAN: A Multi-Granularity Benchmark for Grounded Personality Reasoning
🔥 News
[2026-05] MM-OCEAN was featured as the #2 Paper of the Day on Hugging Face Daily Papers!
Overview
MM-OCEAN is the first benchmark for Grounded Personality Reasoning (GPR), requiring Multimodal Large Language Models (MLLMs) to ground Big Five personality judgments in observable behavioral evidence. The benchmark evaluates models through three tasks of increasing…