Semanlink - Knowledge ; AI@Google AND Retriever-Reader

Βρες μου:

Search Doc:

Προτιμήσεις...

Ετικέτες

Arxiv Doc - Entity Retrieval - Knowledge-augmented Language Models - Neural Search - NLP@Facebook - NLP@Google - Not Encoding Factual Knowledge in Language Model - Retriever-Reader -

Knowledge ; AI@Google AND Retriever-Reader

Common descendants

2 Έγγραφα

[2002.08909] REALM: Retrieval-Augmented Language Model Pre-Training

Tags:

**Augment language model pre-training with a retriever module**, which
is trained using the masked language modeling objective.

> To capture knowledge in a more modular and interpretable way, we augment language model pre-training with a latent knowledge retriever, which allows the model to retrieve and attend over documents from a large corpus such as Wikipedia, used during pre-training, fine-tuning and inference. **For the first time, we show how to pre-train such a knowledge retriever in an unsupervised manner**, using masked language modeling as the learning signal and backpropagating through a retrieval step that considers millions of documents

Hum, #TODO: parallel to be drawn with techniques in [KG-augmented Language Models](tag:knowledge_graph_augmented_language_models) which focus "on the problem of capturing declarative knowledge in the learned parameters of a language model."

[Google AI Blog Post](doc:2020/08/google_ai_blog_realm_integrat)

[Summary](https://joeddav.github.io/blog/2020/03/03/REALM.html) for the [Hugging Face awesome-papers reading group](doc:2021/03/huggingface_awesome_papers_pap)

2020-12-12 About

[1911.03814] Scalable Zero-shot Entity Linking with Dense Entity Retrieval

Tags:

2020-05-02 About