Add LOTTE

[Samoed]

LoTTE Long-Tail Topic-stratified Evaluation for IR that features 12 domain-specific search tests, spanning StackExchange communities and using queries from GooAQ. LoTTE focuses on relatively long-tail topics in its passages, unlike the Open-QA tests and many of the BEIR tasks, and evaluates models on their capacity to answer natural search queries with a practical inten

LoTTE, a new resource for out-of-domain evaluation of retrievers. LoTTE focuses on natural information-seeking queries over long-tail topics, an important yet understudied application space.

Introduced in ColBert2 paper https://arxiv.org/abs/2112.01488
Data https://github.com/stanford-futuredata/ColBERT/blob/main/LoTTE.md