Skip to content

ALLIES

Directory actions

More options

Directory actions

More options

Latest commit

 

History

History

ALLIES

README.md

ALLIES

The code for our paper ALLIES: Prompting Large Language Model with Beam Search.

model

Dataset

NQ

dataset/nq-test.jsonl

TriviaQA

dataset/tqa-test.jsonl

WebQ

dataset/webq-test.jsonl

Released Resources

We release the preprocessed data and trained ckpts in Azure Blob. Here we also provide the file list under this URL:

Click here to see the file list. 
INFO: nq/de-checkpoint-10000/passage_embedding.pb;  Content Length: 60.13 GiB
INFO: nq/de-checkpoint-10000/passage_embedding2id.pb;  Content Length: 160.33 MiB
INFO: webq/de-checkpoint-400/passage_embedding.pb;  Content Length: 60.13 GiB
INFO: webq/de-checkpoint-400/passage_embedding2id.pb;  Content Length: 160.33 MiB
INFO: tq/de-checkpoint-10000/passage_embedding.pb;  Content Length: 60.13 GiB
INFO: tq/de-checkpoint-10000/passage_embedding2id.pb;  Content Length: 160.33 MiB

To download the files, please refer to HOW_TO_DOWNLOAD.

Run

Directly Answer

python main.py --dataset $dataset --task answer_without_retrieval  --apikey $ID

Answer with retrieval

python main.py --dataset $dataset --task answer_with_retrieval --topK $retrieval_num  --apikey $ID

GenRead

python main.py --dataset $dataset --task genread --apikey $ID

Allies

##GENREAD
python main.py --dataset $dataset --task ALLIES --retrieval_type generate --beam_size $beam_size --beam_Depth $beam_depth --ask_question_num $ask_question_num --apikey $ID

##Retrieval
python main.py --dataset $dataset --task ALLIES --topK $retrieval_num --retrieval_type retrieve --beam_size $beam_size --beam_Depth $beam_depth --ask_question_num $ask_question_num --apikey $ID

Parameters

  • $dataset: Dataset for testing
  • $ID: The key for API
  • $beam_size: Beam size
  • $beam_depth: Beam depth
  • $ask_question_num: Ask question number
  • $retrieval_num: Retrieval doc num