| license: mit | |
| Here is the test data of the TruthHypo benchmark (IJCAI 25), which is designed to evaluate the capabilities of LLMs in generating truthful scientific hypothesis. The paper is available at https://arxiv.org/abs/2505.14599. |
| license: mit | |
| Here is the test data of the TruthHypo benchmark (IJCAI 25), which is designed to evaluate the capabilities of LLMs in generating truthful scientific hypothesis. The paper is available at https://arxiv.org/abs/2505.14599. |