Papers › REBUS: A Robust Evaluation Benchmark of Understanding Symbols
REBUS: A Robust Evaluation Benchmark of Understanding Symbols
Andrew Gritsevskiy, Arjun Panickssery, Aaron Kirtland, Derik Kauffman, Hans Gundlach, Irina Gritsevskaya, Joe Cavanagh, Jonathan Chiang, Lydia La Roux, Michelle Hung
We propose a new benchmark evaluating the performance of multimodal large language models on rebus puzzles. The dataset covers 333 original examples of image-based wordplay, cluing 13 categories such as movies, composers, major cities, and food. To achieve good performance on the benchmark of identifying the clued word or phrase, models must combine image recognition and string manipulation with hypothesis testing, multi-step reasoning, and an understanding of human cognition, making for a complex, multimodal evaluation of capabilities. We find that GPT-4o significantly outperforms all other models, followed by proprietary models outperforming all other evaluated models. However, even the best model has a final accuracy of only 42\%, which goes down to just 7\% on hard puzzles, highlighting the need for substantial improvements in reasoning. Further, models rarely understand all parts of a puzzle, and are almost always incapable of retroactively explaining the correct answer. Our benchmark can therefore be used to identify major shortcomings in the knowledge and reasoning of multimodal large language models.
In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Multimodal Reasoning | REBUS | GPT-4V | Accuracy | 24.0 | #1 of 8 | Archive leaderboard | report |
| Multimodal Reasoning | REBUS | Gemini Pro | Accuracy | 13.2 | #2 of 8 | Archive leaderboard | report |
| Multimodal Reasoning | REBUS | LLaVa-1.5-13B | Accuracy | 1.8 | #3 of 8 | Archive leaderboard | report |
| Multimodal Reasoning | REBUS | LLaVa-1.5-7B | Accuracy | 1.5 | #4 of 8 | Archive leaderboard | report |
| Multimodal Reasoning | REBUS | BLIP2-FLAN-T5-XXL | Accuracy | 0.9 | #5 of 8 | Archive leaderboard | report |
| Multimodal Reasoning | REBUS | CogVLM | Accuracy | 0.9 | #6 of 8 | Archive leaderboard | report |
| Multimodal Reasoning | REBUS | QWEN | Accuracy | 0.9 | #7 of 8 | Archive leaderboard | report |
| Multimodal Reasoning | REBUS | InstructBLIP | Accuracy | 0.6 | #8 of 8 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections