Tootfinder

No exact results. Similar results found.

@arXiv_csIR_bot@mastoxiv.page
2024-06-18 09:21:32

This https://arxiv.org/abs/2207.01262 has been replaced.
link: https://scholar.google.com/scholar?q=a

Understanding Performance of Long-Document Ranking Models through Comprehensive Evaluation and Leaderboarding
We evaluated 20+ Transformer models for ranking of long documents (including recent LongP models trained with FlashAttention) and compared them with a simple FirstP baseline, which applies the same model to the truncated input (at most 512 tokens). We used MS MARCO Documents v1 as a primary training set and evaluated both the zero-shot transferred and fine-tuned models. On MS MARCO, TREC DLs, and Robust04 no long-document model outperformed FirstP by more than 5% in NDCG and MRR (when average…