ielab GPT_Ranker issues

ielab / GPT_Ranker

Use GPT-2/T5 to beat traditional LM

2 stars 0 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Run extra experiments

#20 hanglics closed 4 years ago
1
Fine-tune T5 on documents

#19 ArvinZhuang closed 3 years ago
0
Use docTTTTTquery to append 40 samples to each passages and redo reranking task

#18 ArvinZhuang closed 4 years ago
0
Check queries that BM25 outperms our model.

#17 ArvinZhuang closed 4 years ago
0
Pretrain T5 on MS MARCO collection an then finetune on traning set

#16 ArvinZhuang closed 3 years ago
0
Training methods

#15 hanglics closed 3 years ago
0
Alternative models

#14 hanglics closed 3 years ago
0
Retrain models with lowercased documents

#13 ArvinZhuang closed 3 years ago
0
Check the bm25 top 10 is a subset of t5 top 10

#12 hanglics closed 4 years ago
1
Check passage bm25 initial retrieval for not retrieving the relevant document

#11 hanglics closed 4 years ago
1
Document retrieval task

#10 hanglics closed 4 years ago
0
Implement sliding window for MSMARCO Doc

#9 hanglics closed 4 years ago
1
Use T5 to rerank passage devset and testset queries for leaderboard submission.

#8 ArvinZhuang closed 4 years ago
0
MSMARCO Document Train Dataset

#7 hanglics closed 4 years ago
1
Doc2query

#6 ArvinZhuang closed 4 years ago
0
Using Jimmy Lin docTTTTTquery model and feed into our GPT-Ranker

#5 hanglics closed 4 years ago
0
GPT-Ranker using Jimmy Lin Anserini BM25 + doc2query res file to rerank

#4 hanglics closed 4 years ago
1
Fine-tune GPT2 on MSMARCO passages ranking training set.

#3 ArvinZhuang closed 3 years ago
1
Fast GPT inference

#2 ArvinZhuang closed 4 years ago
1
Discuss MS MARCO leaderboard evaluation

#1 ArvinZhuang closed 4 years ago
1