LongBench v2 and LongBench (ACL 2024). Contribute to THUDM/LongBench development by creating an account on GitHub.
LongBench v2 and LongBench (ACL 2024). Contribute to THUDM/LongBench development by creating an account on GitHub.
LongBench v2 and LongBench (ACL 2024). Contribute to THUDM/LongBench development by creating an account on GitHub.
RepoBench-PGiven code in multiple files within a GitHub repository (including cross-file dependencies), predict the next line of code Task construction Note: For all tasks constructed from existing datasets, we use data from the validation or test set of the existing dataset (except for VCSUM)...
LongBench v2 and LongBench (ACL 2024). Contribute to THUDM/LongBench development by creating an account on GitHub.
LongBench v2 and LongBench (ACL 2024). Contribute to THUDM/LongBench development by creating an account on GitHub.
LongBench v2 and LongBench (ACL 2024). Contribute to THUDM/LongBench development by creating an account on GitHub.
LongBench v2的推出不仅为大语言模型的评估提供了新的工具，也为未来的研究指明了方向，强调了提升模型自身理解和推理能力的重要性。智源研究院和腾讯的合作，标志着在AI技术领域的进一步发展，期待这一基准测试能够推动长文本理解和推理技术的进步。主页:https://longbench2.github.io 论文:https://arxiv.org/abs/...
使用单个24G显卡,从0开始训练LLM. Contribute to hongdangshao/baby-llama2-chinese_fix development by creating an account on GitHub.
YuhanLiu11 commented Feb 16, 2025 • edited by github-actions bot This PR adds the LongBench dataset to benchmark_serving.py. Compared to current datasets in benchmark_serving.py, it has much longer input lengths. For example, a dataset from LongBench, NarrativeQA, has 29869 input tokens...