Semanlink - Neural networks ; Embeddings in NLP AND Arxiv Doc

> the problem of building text classifiers with little or no training data.
>
> In recent years, an approach based on neural textual entailment models has been found to give strong results on a diverse range of tasks.

(cf. #[NLI](tag:nli), using the input text as the premise and the text representing the label as the hypothesis)

> In this work, we show that **with proper pre-training, Siamese Networks that embed texts and labels** offer a competitive alternative.
>
> We introduce **label tuning: fine-tuning the label embeddings only**. While giving lower performance than model fine-tuning (which updates all params of the model), this approach has the architectural advantage that a single encoder can be shared by many different tasks (we only fine-tune the label embeddings)
> The drop in quality can
be compensated by using a variant of **[Knowledge distillation](tag:knowledge_distillation)**

[Github](https://tinyurl.com/label-tuning), [Tweet](doc:2022/03/thomas_muller_sur_twitter_pa)

2022-03-30 About

[2004.09813] Making Monolingual Sentence Embeddings Multilingual using Knowledge Distillation

Tags:

2022-03-18 About

[2110.10778] Contrastive Document Representation Learning with Graph Attention Networks

Tags:

2022-03-10 About

[2109.06304] Phrase-BERT: Improved Phrase Embeddings from BERT with an Application to Corpus Exploration

Tags:

2022-02-25 About

[2004.07180] SPECTER: Document-level Representation Learning using Citation-informed Transformers

Tags:

2022-01-29 About

[2112.07577] GPL: Generative Pseudo Labeling for Unsupervised Domain Adaptation of Dense Retrieval

Tags:

An unsupervised domain adaptation technique for dense retrieval models

1. synthetic queries
are generated for each passage from the target corpus (using an existing pre-trained [T5](tag:text_to_text_transfer_transformer)
encoder-decoder)
2. the generated queries are used for mining negative
passages (retrieving the most similar
paragraphs using an existing dense retrieval
model == hard negatives!)
3. the query-passage pairs are labeled by a cross-encoder and used to train the domain-adapted
dense retriever (using method described in [Hofstätter et al.,
2020](doc:2021/12/2010_02666_improving_efficien))

[Nils Reimers sur Twitter](doc:2021/12/nils_reimers_sur_twitter_do_), [GitHub](https://github.com/UKPLab/gpl),  by the author of [TSDAE](doc:2021/09/2104_06979_tsdae_using_trans)

Claims to improve "Doc2Query" [Document Expansion by Query Prediction](doc:2022/01/1904_08375_document_expansion): ([src](https://twitter.com/KexinWang2049/status/1471435779415150598))

> - GPL: Uses doc2query to construct synthetic data and does knowledge distillation (i.e. training) on that data.
> - Doc2query: Generates queries to extend the documents and use BM25 on top of them w/o training.

2021-12-15 About

[2109.08133] Phrase Retrieval Learns Passage Retrieval, Too

Tags:

2021-09-30 About

[2104.06979] TSDAE: Using Transformer-based Sequential Denoising Auto-Encoder for Unsupervised Sentence Embedding Learning

Tags:

> The most
successful previous approaches like InferSent (Conneau
et al., 2017), Universial Sentence Encoder
(USE) (Cer et al., 2018) and SBERT (Reimers and
Gurevych, 2019) heavily relied on labeled data to
train sentence embedding models.
>
> TSDAE can
achieve up to 93.1% of the performance of indomain
supervised approaches. Further, we
show that TSDAE is **a strong domain adaptation
and pre-training method for sentence
embeddings**, significantly outperforming other
approaches like Masked Language Model.

> During training, TSDAE
encodes corrupted sentences into fixed-sized
vectors and requires the decoder to reconstruct the
original sentences from this sentence embedding.

- <https://www.sbert.net/examples/unsupervised_learning/TSDAE/README.html>
- [github](https://github.com/UKPLab/sentence-transformers/tree/master/examples/unsupervised_learning/TSDAE)
- [UKPLab/sentence-transformers: Sentence Embeddings with BERT & XLNet](doc:2020/07/ukplab_sentence_transformers_s)
- [twitter](https://twitter.com/KexinWang2049/status/1433361957579538432):

> **TSDAE can learn domain-specific sentence embeddings with unlabeled sentences**
>
> Most importantly, instead of STS (Semantic Textual Similarity), **we suggest evaluating unsupervised sentence embeddings on the domain-specific tasks&datasets, which is the real use case for them**. Actually, STS scores do not correlate with performance on specific tasks.

2021-09-01 About

[2011.05864] On the Sentence Embeddings from Pre-trained Language Models

Tags:

2021-04-19 About

[1909.01259] Neural Attentive Bag-of-Entities Model for Text Classification

Tags:

A model that performs **text classification using entities in a knowledge base**.

> Entities provide unambiguous and relevant semantic signals that are beneficial for capturing semantics in texts. We combine **simple high-recall entity detection based on a dictionary** (word->list of entities), to detect entities in a document, with a novel neural **attention mechanism that enables the model to focus on a small number of unambiguous and relevant entities**.

2 steps:

1. Entity detection
2. Classification using the detected entities (+text) as inputs

Regarding entity linking, a local model which uses cosine
similarity between the embedding of the target
entity and the word-based representation of
the document to capture the relevance of an entity
given a document.

Embeddings from the KB: computed using [#Wikipedia2Vec](tag:wikipedia2vec) (similar words and entities
close to one another in a unified vector space)

Model using attention, with 2 features :

- cosine similarity between the
embedding of the entity and the word based
representation of the document
- the probability that the entity
name refers to the entity in KB.

Somewhat [related](doc:2020/01/investigating_entity_knowledge_)

### Conclusion:

>a neural
network model that performs text classification using
entities in Wikipedia. We combined simple
dictionary-based entity detection with a neural attention
mechanism to enable the model to focus
on a small number of unambiguous and relevant
entities in a document.

2020-09-02 About

[1812.06280] Wikipedia2Vec: An Efficient Toolkit for Learning and Visualizing the Embeddings of Words and Entities from Wikipedia

Tags:

2020-09-02 About

[1910.00163] Specializing Word Embeddings (for Parsing) by Information Bottleneck

Tags:

2020-06-29 About

[1709.03933] Hash Embeddings for Efficient Word Representations

Tags:

2020-05-19 About

[1910.04126] Scalable Nearest Neighbor Search for Optimal Transport

Tags:

2020-02-20 About

[1908.10084] Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks

Tags:

2019-08-28 About

[1810.04882] Towards Understanding Linear Word Analogies

Tags:

2019-06-24 About

[1906.02715] Visualizing and Measuring the Geometry of BERT

Tags:

2019-06-07 About

[1703.02507] Unsupervised Learning of Sentence Embeddings using Compositional n-Gram Features

Tags:

2019-03-25 About

[1902.11269] Efficient Contextual Representation Learning Without Softmax Layer

Tags:

**how to accelerate contextual representation learning**.

> Contextual representation models are difficult to train due to the large parameter sizes and high computational complexity

> We find that the softmax layer (the output layer) causes significant inefficiency due to the large vocabulary size.
Therefore, we redesign the learning objectiv.
> Specifically, the proposed approach bypasses the softmax layer by performing language modeling with dimension reduction, and allows the models to leverage pre-trained word embeddings.
Our framework reduces the time spent on the output layer to a negligible level, eliminates almost all the trainable parameters of the softmax layer and performs language modeling without truncating the vocabulary.
When applied to ELMo, our method achieves a 4 times speedup and eliminates 80% trainable parameters while achieving competitive performance on downstream tasks.

**decouples learning contexts and words**

> Instead of using
a softmax layer to predict the distribution of the
missing word, we utilize and extend the SEMFIT
layer (Kumar and Tsvetkov, 2018) to **predict the
embedding of the missing word**.

2019-03-02 About

[1902.10618] Still a Pain in the Neck: Evaluating Text Representations on Lexical Composition

Tags:

2019-02-28 About

[1601.01343] Joint Learning of the Embedding of Words and Entities for Named Entity Disambiguation

Tags:

2019-01-27 About

[1811.05370] Unsupervised Transfer Learning for Spoken Language Understanding in Intelligent Agents

Tags:

2018-11-20 About

[1710.06632] Towards a Seamless Integration of Word Senses into Downstream NLP Applications

Tags:

2018-10-09 About

[1601.03764] Linear Algebraic Structure of Word Senses, with Applications to Polysemy

Tags:

> Here it is shown that multiple word senses reside
in linear superposition within the word
embedding and simple sparse coding can recover
vectors that approximately capture the
senses

> Each extracted word sense is accompanied by one of about  2000 “discourse atoms” that gives a succinct description of which other words co-occur with that word sense.

> The success of the approach is mathematically explained using a variant of
the random walk on discourses model

("random walk": a generative model for language). Under the assumptions of this model,  there
exists a linear relationship between the vector of a
word w and the vectors of the words in its contexts (It is not the average of the words in w's context, but in a given corpus the matrix of the linear relationship does not depend on w. It can be estimated, and so we can compute the embedding of a word from the contexts it belongs to)

[Related blog post](/doc/?uri=https%3A%2F%2Fwww.offconvex.org%2F2016%2F07%2F10%2Fembeddingspolysemy%2F)

2018-08-28 About

[1608.05426] A Strong Baseline for Learning Cross-Lingual Word Embeddings from Sentence Alignments

Tags:

2018-07-23 About

[1806.06259] Evaluation of sentence embeddings in downstream and linguistic probing tasks

Tags: