Semanlink - [2012.04584] Distilling Knowledge from Reader to Retriever for Question Answering

[2012.04584] Distilling Knowledge from Reader to Retriever for Question Answering

Tags:

> a method to train an information retrieval module for downstream tasks, **without using pairs of queries and documents as annotations**.

Uses two models (standard pipeline for open-domain QA):

- the first one retrieves documents from a large source of knowledge (the retriever)
- the second one processes the support documents to solve the task (the reader).

> First the retriever selects support passages in a large knowledge
source. Then these passages are processed by the reader, along with the question, to generate an
answer

Inspired by knowledge distillation: the reader model is the teacher and the retriever is the student.

> More precisely, we use a sequence-to-sequence model as the reader, and use
the attention activations over the input documents as synthetic labels to train the retriever. 
> (**train the retriever by learning to approximate the attention score of the reader**)

Refers to:

- [REALM: Retrieval-Augmented Language Model Pre-Training](doc:2020/12/2002_08909_realm_retrieval_a)
- [Dehghani: Neural Ranking Models with Weak Supervision](doc:?uri=https%3A%2F%2Farxiv.org%2Fabs%2F1704.08803)

About This Document

sl:arxiv_author :
- Edouard Grave
- Gautier Izacard
sl:arxiv_firstAuthor : Gautier Izacard
sl:arxiv_num : 2012.04584
sl:arxiv_published : 2020-12-08T17:36:34Z
sl:arxiv_summary : The task of information retrieval is an important component of many natural language processing systems, such as open domain question answering. While traditional methods were based on hand-crafted features, continuous representations based on neural networks recently obtained competitive results. A challenge of using such methods is to obtain supervised data to train the retriever model, corresponding to pairs of query and support documents. In this paper, we propose a technique to learn retriever models for downstream tasks, inspired by knowledge distillation, and which does not require annotated pairs of query and documents. Our approach leverages attention scores of a reader model, used to solve the task based on retrieved documents, to obtain synthetic labels for the retriever. We evaluate our method on question answering, obtaining state-of-the-art results.@en
sl:arxiv_title : Distilling Knowledge from Reader to Retriever for Question Answering@en
sl:arxiv_updated : 2020-12-08T17:36:34Z
sl:bookmarkOf : https://arxiv.org/abs/2012.04584
sl:creationDate : 2020-12-11
sl:creationTime : 2020-12-11T16:48:13Z
sl:relatedDoc :
- https://arxiv.org/abs/1704.08803
- http://www.semanlink.net/doc/2020/12/2002_08909_realm_retrieval_a

File info

Bookmark of: https://arxiv.org/abs/2012.04584

Linked From

[2206.10658] Questions Are All You Need to Train a Dense Passage Retriever

Tags:

> **approach for training dense retrieval models that does not require any labeled training data**. Dense retrieval is a central challenge for open-domain tasks, such as Open QA, where state-of-the-art methods typically require large supervised datasets with custom hard-negative mining and denoising of positive examples.
>
> ART, in contrast, only requires access to unpaired inputs and outputs (e.g. questions and potential answer documents).
>
> It uses a new document-retrieval autoencoding scheme, where
> 1. an input question is used to retrieve a set of evidence documents, and
> 2. the documents are then used to compute the probability of reconstructing the original question.
>
> Training for retrieval based on question reconstruction enables effective unsupervised learning of both document and question encoders, which can be later incorporated into complete Open QA systems without any further finetuning.

[Tweet](doc:2022/07/devendra_singh_sachan_sur_twitt)

> Given an
input question, ART first retrieves a small set
of possible evidences documents. It then recon
structs
the original question by attending to these
documents
>
> The
key idea in ART is to consider the retrieved documents
as a noisy representation of the original
question and question reconstruction probability
as a way of denoising that provides soft-labels for
how likely each document is to have been the correct
result

Refers to [[IZACARD 2012.04584] Distilling Knowledge from Reader to Retriever for Question Answering](doc:2020/12/2012_04584_distilling_knowled)

2022-07-06 About

Documents with similar tags (experimental)

[2002.06275] TwinBERT: Distilling Knowledge to Twin-Structured BERT Models for Efficient Retrieval

Tags:

2023-08-27 About

[2009.13013] SPARTA: Efficient Open-Domain Question Answering via Sparse Transformer Matching Retrieval

Tags:

2023-04-06 About

[1904.04458] Knowledge-Augmented Language Model and its Application to Unsupervised Named-Entity Recognition

Tags:

Knowledge Augmented
Language Model (KALM)

a language
model with access to information available in a
KB, no assumptions
about the availability of additional components
(such as Named Entity Taggers) or annotations

> While classes of
named entities (e.g., person or location) occur frequently,
each individual name (e.g, Atherton or
Zhouzhuang) may be observed infrequently even
in a very large corpus of text. As a result language
models learn to represent accurately only the most
popular named entities

> knowing that Alice is a name
used to refer to a person should give ample information
about the context in which the word may
occur (e.g., Bob visited Alice).

> ---

> extends a traditional **RNN LM**

> we enhance a traditional LM with a
gating mechanism that controls whether a particular
word is modeled as a general word or as a reference
to an entity
>
> We train the model end-to-end
with only the traditional predictive language modeling
perplexity objective
>
> KALM is trained end-to-end using
a predictive objective on large corpus of text.

> To the best of our knowledge, KALM is the first
unsupervised neural NER approach.

> KALM extends a traditional, RNN-based neural
LM.

2022-08-31 About

[2012.15156] A Memory Efficient Baseline for Open Domain Question Answering

Tags:

2022-08-08 About

[2208.03299] Few-shot Learning with Retrieval Augmented Language Model

Tags:

2022-08-08 About

[2207.06300] Re2G: Retrieve, Rerank, Generate

Tags:

2022-07-14 About

[2206.10658] Questions Are All You Need to Train a Dense Passage Retriever

Tags:

[Tweet](doc:2022/07/devendra_singh_sachan_sur_twitt)

Refers to [[IZACARD 2012.04584] Distilling Knowledge from Reader to Retriever for Question Answering](doc:2020/12/2012_04584_distilling_knowled)

2022-07-06 About

[2112.09118] Towards Unsupervised Dense Information Retrieval with Contrastive Learning

Tags:

2021-12-21 About

[2004.04906] Dense Passage Retrieval for Open-Domain Question Answering

Tags:

2021-06-03 About

[2010.02194] Self-training Improves Pre-training for Natural Language Understanding

Tags:

2021-03-12 About

[2010.00904] Autoregressive Entity Retrieval

Tags:

2021-01-14 About

[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

[1911.00172] Generalization through Memorization: Nearest Neighbor Language Models

Tags:

2019-12-20 About

[1909.04120] Span Selection Pre-training for Question Answering

Tags:

> a **new pre-training task inspired by reading
comprehension** and an **effort to avoid encoding general knowledge in the transformer network itself**

Current transformer architectures store general knowledge -> large models, long pre-training time. Better to offload the requirement of general knowledge to a sparsely activated network.

"Span selection" as an additional auxiliary task: the query is a sentence drawn from a corpus
with a term replaced with a special token: [BLANK]. The term replaced by the blank is the answer term. The passage is
relevant as determined by a BM25 search, and answer-bearing (containing the answer
term). Unlike BERT’s cloze task, where the answer must be drawn from the model itself, the answer is found in a passage
using language understanding.

> **We hope to progress to a model of general purpose language modeling that uses an indexed long
term memory to retrieve world knowledge, rather than holding it in the densely activated transformer encoder layers.**

2019-09-18 About