T5 Large Ssm Nq

🚀 Google's T5 for Closed Book Question Answering

This model leverages Google's T5 for Closed Book Question Answering, offering a solution to answer questions without external context. It has been pre - trained and fine - tuned on multiple datasets, showing promising results in question - answering tasks.

🚀 Quick Start

Google's T5 is applied for Closed Book Question Answering. The model undergoes a multi - stage training process. First, it is pre - trained using T5's denoising objective on C4. Then, it is additionally pre - trained using REALM's salient span masking objective on Wikipedia. Finally, it is fine - tuned on Natural Questions (NQ).

Note: The model was fine - tuned on 100% of the train splits of Natural Questions (NQ) for 10k steps.

Other community Checkpoints can be found here.

The related paper is How Much Knowledge Can You Pack Into the Parameters of a Language Model? by Adam Roberts, Colin Raffel, Noam Shazeer.

✨ Features

• Multi - stage Training: Trained on different datasets with various objectives to enhance knowledge storage and retrieval.
• Closed - book Question Answering: Capable of answering questions without relying on external knowledge sources.

📦 Installation

No specific installation steps are provided in the original document, so this section is skipped.

💻 Usage Examples

Basic Usage

The model can be used as follows for closed book question answering:

from transformers import AutoModelForSeq2SeqLM, AutoTokenizer

t5_qa_model = AutoModelForSeq2SeqLM.from_pretrained("google/t5-large-ssm-nq")
t5_tok = AutoTokenizer.from_pretrained("google/t5-large-ssm-nq")

input_ids = t5_tok("When was Franklin D. Roosevelt born?", return_tensors="pt").input_ids
gen_output = t5_qa_model.generate(input_ids)[0]

print(t5_tok.decode(gen_output, skip_special_tokens=True))

# should give "December 26, 1892" => close, but not correct.

📚 Documentation

Results on Natural Questions - Test Set

Abstract

It has recently been observed that neural language models trained on unstructured text can implicitly store and retrieve knowledge using natural language queries. In this short paper, we measure the practical utility of this approach by fine - tuning pre - trained models to answer questions without access to any external context or knowledge. We show that this approach scales with model size and performs competitively with open - domain systems that explicitly retrieve answers from an external knowledge source when answering questions. To facilitate reproducibility and future work, we release our code and trained models at https://goo.gle/t5 - cbqa.

📄 License

This model is released under the [apache - 2.0](https://www.apache.org/licenses/LICENSE - 2.0) license.