Closed-ended Reasoning on AIME24
0.1333AccuracyTF-TTCL
Evaluation Results
| Method | Links | |
|---|---|---|
| TF-TTCLPublication=-, Backbone=Llama-3.1-8B-Instruct2026.04 | 0.1333 | |
| TentPublication=ICLR 2021, Backbone=Llama-3.1-8B-Instruct2026.04 | 0.1 | |
| EATAPublication=ICML 2022, Backbone=Llama-3.1-8B-Instruct2026.04 | 0.0667 | |
| COMEPublication=ICLR 2025, Backbone=Llama-3.1-8B-Instruct2026.04 | 0.0667 | |
| TLMPublication=ICML 2025, Backbone=Llama-3.1-8B-Instruct2026.04 | 0.0667 | |
| Base LLMPublication=-, Backbone=Llama-3.1-8B-Instruct, zero-shot=true2026.04 | 0.0333 | |
| TF-GRPOPublication=arXiv 2025, Backbone=Llama-3.1-8B-Instruct2026.04 | 0.0333 |