TREC DL
A classic evaluation in the information retrieval field, providing a large-scale document collection and human relevance judgments, with emphasis on evaluating the performance of deep models in passage-level reranking.
Timeline (last 90 days)
DecisionRank achieves state-of-the-art performance on general and reasoning-intensive retrieval benchmarks, including TREC DL, BEIR, and BRIGHT.
The 8B-scale DecisionRank outperforms its generative listwise counterpart while running 24.5 times faster on TREC DL and 16.7 times faster on BRIGHT.
All Facts (2)
DecisionRank achieves state-of-the-art performance on general and reasoning-intensive retrieval benchmarks, including TREC DL, BEIR, and BRIGHT.
50%UnverifiedThe 8B-scale DecisionRank outperforms its generative listwise counterpart while running 24.5 times faster on TREC DL and 16.7 times faster on BRIGHT.
50%