| MultiArith | | Accuracy100 | | 324 | 1mo ago |
| GSM8K | DUP | Accuracy97.1 | | 272 | 1mo ago |
| GSM8K (test) | SGE | Accuracy97.35 | | 203 | 29d ago |
| AddSub | CLoT | Accuracy99 | | 149 | 1mo ago |
| MultiArith (test) | LoRAALL | Accuracy99.5 | | 136 | 29d ago |
| Arithmetics | | Accuracy100 | | 106 | 3mo ago |
| SVAMP | Graph-GRPO | Accuracy96.01 | | 87 | 1mo ago |
| SVAMP (test) | SGE | Accuracy98.16 | | 84 | 29d ago |
| SingleEq | PAL | Accuracy98.8 | | 73 | 1mo ago |
| AQuA (test) | SGE | Accuracy74.63 | | 72 | 29d ago |
| ASDiv | | Accuracy93.74 | | 69 | 1mo ago |
| AQUA | DUP | Accuracy77.1 | | 57 | 1mo ago |
| SVAMP | Automatic Model Selection with LLMs | Accuracy (Overall)93.7 | | 54 | 4mo ago |
| In-domain (test) | | Accuracy53.4 | | 50 | 4mo ago |
| SVAMP, GSM8K, AddSub, MultiArith, AQUA, SingleEq | DUP | Average Score92.9 | | 46 | 3mo ago |
| MATH | LLM-PeerReview-W | Accuracy71 | | 39 | 2mo ago |
| AQuA, GSM8K, MAWPS, SVAMP | Qwen2.5-14B | AQuA Accuracy62.2 | | 31 | 4mo ago |
| GSM8K sampled subset n=200 (test) | Chain-of-Thought Prompting | Accuracy95.5 | | 30 | 4mo ago |
| Game of 24 (test) | ICRL Preset | Success Rate90 | | 28 | 2mo ago |
| Average Arithmetic Reasoning Tasks | | Accuracy84.3 | | 26 | 1mo ago |
| Countdown | GRPO + coupled rhythm credit | Accuracy63.1 | | 25 | 1mo ago |
| Arithmetic Reasoning Benchmarks (MultiArith, GSM8K, AddSub, AQuA, SingleEQ, SVAMP, MAWPS) MATH-10K fine-tuned (test) | S2FT | MultiArith Accuracy99.67 | | 24 | 3mo ago |
| MAWPS | | Accuracy95.94 | | 23 | 1mo ago |
| AddSub (test) | CoT + Skill-Based | Accuracy96.71 | | 22 | 29d ago |
| Matchstick Arithmetic (held-out) | CORE | Accuracy90.7 | | 18 | 1mo ago |