Buckets:

---

SmallSquirrelaaaaaaa

/

LLM-Uncertainty-Bench-bucket

Datasets comprising 10,000 instances each used for uncertainty quantification in LLMs.

1. mmlu_10k is used for question answering.
2. cosmosqa_10k is used for reading comprehension.
3. hellaswag_10k is used for commonsense inference.
4. halu_dialogue is used for dialogue response selection.
5. halu_summarization is used for document summarization.

For more details on how these datasets are utilized, check out our github repo: https://github.com/smartyfh/LLM-Uncertainty-Bench/tree/main