together-search-bench

## Together-Search-Bench Dataset This dataset is used for delopment and evaluations for [Together Open Deep Research](https://github.com/togethercomputer/open_deep_research). The data is composed of 50 samples from each of the following datasets: simpleqa: [basicv8vc/SimpleQA](https://huggingface.co/datasets/basicv8vc/SimpleQA) frames: [google/frames-benchmark](https://huggingface.co/datasets/google/frames-benchmark) hotpotqa: [hotpotqa/hotpot_qa](https://huggingface.co/datasets/hotpotqa/hotpot_qa) ### License Information Part of the data derived from [hotpotqa](https://huggingface.co/datasets/hotpotqa/hotpot_qa) by Yang et al., licensed under [CC BY-SA 4.0](https://creativecommons.org/licenses/by-sa/4.0/legalcode). Modifications include the full dataset subsampling. Part of the data derived from [frames](https://huggingface.co/datasets/google/frames-benchmark) by Krishna et al., licensed under [Apache License 2.0](https://www.apache.org/licenses/LICENSE-2.0). Modifications include the full dataset subsampling. Part of the data derived from [simpleqa](https://huggingface.co/datasets/basicv8vc/SimpleQA) uploaded to huggingface by `basicv8vc`, licensed under [MIT License](https://opensource.org/license/mit). Modifications include the full dataset subsampling. simpleqa dataset is originally from [OpenAI](https://openai.com/index/introducing-simpleqa/). The distribution of this work should refer to and observe the licensing information descibed above.