OpenCodePapers

open-domain-dialog-on-kilt-wizard-of

Open-Domain Dialog
Dataset Link
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
PaperCodeKILT-RLR-PrecRecall@5ROUGE-LF1KILT-F1ModelNameReleaseDate
[]()11.9256.0874.2717.0619.1913.39Hindsight
Re2G: Retrieve, Rerank, Generate✓ Link11.3960.179.9816.7618.912.98Re2G2022-07-13
[]()10.4557.5578.9616.6518.3411.63intersect
[]()10.3655.3778.4516.3618.5711.79KGI
[]()7.5957.7574.6111.5713.118.75RAG
[]()6.5541.5468.2513.9415.667.57Wikipedia
[]()5.9141.0667.1313.2715.126.96Multitask DPR + BART
[]()4.4139.0651.6311.4212.154.8Routing Transformer, c-REALM
[]()3.7125.4651.1913.2315.194.37BART + DPR
[]()2.0455.7175.592.923.092.18multitask
[]()1.8518.3518.3510.1111.852.2TransMemNet
[]()0.064.7982.150.00.00.0chriskuei
[]()0.062.8877.740.00.00.0GENRE
[]()0.059.1169.10.00.00.0TABi
[]()0.041.0667.130.00.00.0Multi-task DPR
[]()0.00.00.015.9317.30.0aa_evalai
[]()0.00.00.015.7117.280.0Sphere
[]()0.00.00.013.3514.820.0bart-base
[]()0.00.00.012.8113.750.0multi-task small
KILT: a Benchmark for Knowledge Intensive Language Tasks✓ Link0.00.00.012.413.530.0T5-base2020-09-04
[]()0.00.00.011.7712.860.0BART