doc2query-viT5

This is a doc2query model based on viT5

It can be used for:

• Document expansion: You generate for your paragraphs 20-40 queries and index the paragraphs and the generates queries in a standard BM25 index like Elasticsearch, OpenSearch, or Lucene. The generated queries help to close the lexical gap of lexical search, as the generate queries contain synonyms. Further, it re-weights words giving important words a higher weight even if they appear seldomn in a paragraph.
• Domain Specific Training Data Generation: It can be used to generate training data to learn an embedding model.

Usage

from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
import torch

model_name = 'r1ck/doc2query-viT5'
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)

text = "Chiến tranh Lạnh (1947-1991, tiếng Anh: Cold War) là chỉ đến sự căng thẳng địa chính trị và xung đột ý thức hệ đỉnh điểm giữa hai siêu cường (đứng đầu và đại diện hai khối đối lập): Hoa Kỳ (chủ nghĩa tư bản) và Liên Xô (chủ nghĩa xã hội)."


def create_queries(para):
    input_ids = tokenizer.encode(para, return_tensors='pt')
    with torch.no_grad():
        # Here we use top_k / top_k random sampling. It generates more diverse queries, but of lower quality
        sampling_outputs = model.generate(
            input_ids=input_ids,
            max_length=64,
            do_sample=True,
            top_p=0.95,
            top_k=10, 
            num_return_sequences=5
            )
        
        # Here we use Beam-search. It generates better quality queries, but with less diversity
        beam_outputs = model.generate(
            input_ids=input_ids, 
            max_length=64, 
            num_beams=5, 
            no_repeat_ngram_size=2, 
            num_return_sequences=5, 
            early_stopping=True
        )


    print("Paragraph:")
    print(para)
    
    print("\nBeam Outputs:")
    for i in range(len(beam_outputs)):
        query = tokenizer.decode(beam_outputs[i], skip_special_tokens=True)
        print(f'{i + 1}: {query}')

    print("\nSampling Outputs:")
    for i in range(len(sampling_outputs)):
        query = tokenizer.decode(sampling_outputs[i], skip_special_tokens=True)
        print(f'{i + 1}: {query}')

create_queries(text)

Note: model.generate() is non-deterministic for top_k/top_n sampling. It produces different queries each time you run it.

Training

This model fine-tuned VietAI/vit5-base on 30k vietnamese passage-question pairs