RankRAG-llama3-70b (Zero-Shot, KILT)

Reported on 2 benchmarks across 1 task · 1 paper · 1 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing2 results

Question AnsweringonTriviaQA
EM· uses extra data· 2024-07-02
86.5
best: 87.5 (Claude 2 (few-shot, k=5))
SOTA
RankRAG: Unifying Context Ranking with Retrieval-Augmented Generation in LLMs arXiv:2407.02485
Question AnsweringonNatural Questions
EM· uses extra data· 2024-07-02
54.2
best: 64 (Atlas (full, Wiki-dec-2018 index))
RankRAG: Unifying Context Ranking with Retrieval-Augmented Generation in LLMs arXiv:2407.02485