Thailand

Pre-trained transformer language models have shown remarkable performance on a variety of NLP tasks. However, recent research has suggested that phrase-level representations in these models reflect heavy influences of lexical content, but lack evidence of sophisticated, compositional phrase information. Here we investigate the impact of fine-tuning on the capacity of contextualized embeddings to capture phrase meaning information beyond lexical content. Specifically, we fine-tune models on an adversarial paraphrase classification task with high lexical overlap, and on a sentiment classification task. After fine-tuning, we analyze phrasal representations in controlled settings following prior work. We find that fine-tuning largely fails to benefit compositionality in these representations, though training on sentiment yields a small, localized benefit for certain models. In follow-up analyses, we identify confounding cues in the paraphrase dataset that may explain the lack of composition benefits from that task, and we discuss potential factors underlying the localized benefits from sentiment training.

IJCNLP-AACL 2021

On the Interplay Between Fine-tuning and Composition in Transformers

composition

computational linguistics

machine learning

transformers

**Welcome to ACL-IJCNLP 2021!**

The great event is jointly organized by the Association for Computational Linguistics (ACL) and Asian Federation of Natural Language Processing (AFNLP). 

As in previous years, the program of the conference includes a poster session, tutorials, workshops and demonstrations in addition to the main conference.


We were able to keep the registration fees similar to those charged for the virtual ACL 2020. The one fee allows attendance at the main conference and any/all tutorials and workshops. These fees would be $125 Regular Early and $175 Regular Late; $50 Student Early and $75 Student Late. Early registration closes at midnight July 11, 2021 (Eastern Daylight Time).

**Reminder:** It is ACL’s policy that at least one author of each accepted paper (including ACL Finding papers) must register for the conference.

**Reminder2:** Underline site will open closer to the event. If you already registered you will receive access detail

Registration is now open

The great event is jointly organized by the Association for Computational Linguistics (ACL) and Asian Federation of Natural Language Processing (AFNLP).

workshop paper

Static word embeddings that represent words by a single vector 
cannot capture the variability of word meaning in different linguistic
and extralinguistic contexts. Building on prior work on contextualized 
and dynamic word embeddings, we introduce dynamic contextualized 
word embeddings that represent words as a function of both linguistic
and extralinguistic context. Based on a pretrained language model (PLM), 
dynamic contextualized word embeddings model time and social space 
jointly, which makes them attractive for a range of NLP tasks involving
semantic variability. We highlight potential application scenarios by means 
of qualitative and quantitative analyses on four English datasets.

Dynamic Contextualized Word Embeddings

While there is a large amount of research in the field of Lexical Semantic Change Detection, only few approaches go beyond a standard benchmark evaluation of existing models. In this paper, we propose a shift of focus from change detection to change discovery, i.e., discovering novel word senses over time from the full corpus vocabulary. By heavily fine-tuning a type-based and a token-based approach on recently published German data, we demonstrate that both models can successfully be applied to discover new words undergoing meaning change. Furthermore, we provide an almost fully automated framework for both evaluation and discovery.

Lexical Semantic Change Discovery

Transformer-based language models have taken many fields in NLP by storm. BERT and its derivatives dominate most of the existing evaluation benchmarks, including those for Word Sense Disambiguation (WSD), thanks to their ability in capturing context-sensitive semantic nuances. However, there is still little knowledge about their capabilities and potential limitations in encoding and recovering word senses. In this article, we provide an in-depth quantitative and qualitative analysis of the celebrated BERT model with respect to lexical ambiguity. One of the main conclusions of our analysis is that BERT can accurately capture high-level sense distinctions, even when a limited number of examples is available for each word sense. Our analysis also reveals that in some cases language models come close to solving coarse-grained noun disambiguation under ideal conditions in terms of availability of training data and computing resources. However, this scenario rarely occurs in real-world settings and, hence, many practical challenges remain even in the coarse-grained setting. We also perform an in-depth comparison of the two main language model based WSD strategies, namely, fine-tuning and feature extraction, finding that the latter approach is more robust with respect to sense bias and it can better exploit limited available training data. In fact, the simple feature extraction strategy of averaging contextualized embeddings proves robust even using only three training sentences per word sense, with minimal improvements obtained by increasing the size of this training data.

Analysis and Evaluation of Language Models for Word Sense Disambiguation

Pruning has been demonstrated as an effective way of reducing computational complexity for deep networks, especially CNNs for computer vision tasks. In this paper, we investigate the opportunity to accelerate the inference of large-scale pre-trained language model via pruning. We propose EBERT, a dynamic structured pruning algorithm for efficient BERT inference. Unlike previous methods that randomly prune the model weights for static inference, EBERT dynamically determines and prunes the unimportant heads in multi-head self-attention layers and the unimportant structured computations in feed-forward network for each input sample at run-time. Experimental results show that our proposed EBERT outperforms other state-of-the-art methods on different tasks.

EBERT: Efficient BERT Inference with Dynamic Structured Pruning

How to combine several pieces of evidence to verify a claim is an interesting semantic task. Very complex methods have been proposed, combining different evidence vectors using an evidence interaction graph. In this paper, we show that in case of inference based on transformer models, two effective approaches use either (i) a simple application of max pooling over the Transformer evidence vectors; or (ii) computing a weighted sum of the evidence vectors. Our experiments on the FEVER claim verification task show that the methods above achieve the state of the art, constituting strong baseline for much more computationally complex methods.

Strong and Light Baseline Models for Fact-Checking Joint Inference

Conventional autoregressive models have achieved great success in text generation but suffer from the exposure bias problem in that token sequences in the training and in the generation stages are mismatched.
While generative adversarial networks (GANs) can remedy this problem, existing implementations of GANs directly on the discrete outputs tend to be unstable and lack diversity.
In this work, we propose TILGAN, a Transformer-based Implicit Latent GAN, which combines a Transformer autoencoder and GAN in the latent space with a novel design and distribution matching based on the Kullback-Leibler (KL) divergence.
Specifically, to improve the local and global coherence, we explicitly introduce a multi-scale discriminator to capture the semantic information at varying scales among the sequence of hidden representations encoded by Transformer. 
Moreover, the decoder is enhanced by an additional KL loss to be consistent with the latent-generator. 
Experimental results on three benchmark datasets demonstrate the validity and effectiveness of our model, by obtaining significant improvements and a better quality-diversity trade-off in automatic and human evaluation for both unconditional and conditional generation tasks. Our code is available at https://github.com/shizhediao/TILGAN.

TILGAN: Transformer-based Implicit Latent GAN for Diverse and Coherent Text Generation

Continuous representations of linguistic structures are an important part of modern natural language processing systems. Despite the diversity, most of the existing log-multilinear embedding models are organized under vector operations. However, these operations can not precisely represent the compositionality of natural language due to a lack of order-preserving properties. In this work, we focus on one of the promising alternatives based on the embedding of documents and words in the rotation group through the generalization of the coupled tensor chain decomposition to the exponential family of the probability distributions. In this model, documents and words are represented as matrices, and n-grams representations are combined from word representations by matrix multiplication. The proposed model is optimized via noise-contrastive estimation. We show empirically that capturing word order and higher-order word interactions allows our model to achieve the best results in several document classification benchmarks.


Documents Representation via Generalized Coupled Tensor Chain with the Rotation Group Constraint

Existing Natural Language Understanding (NLU) models have been shown to incorporate dataset biases leading to strong performance on in-distribution (ID) test sets but poor performance on out-of-distribution (OOD) ones. We introduce a simple yet effective debiasing framework whereby the shallow representations of the main model are used to derive a bias model and both models are trained simultaneously. We demonstrate on three well-studied NLU tasks that despite its simplicity, our method leads to competitive OOD results. It significantly outperforms other debiasing approaches on two tasks, while still delivering high in-distribution performance.


End-to-End Self-Debiasing Framework for Robust NLU Training

We consider a joint information extraction (IE) model, solving named entity recognition, coreference resolution and relation extraction jointly over the whole document. In particular, we study how to inject information from a knowledge base (KB) in such IE model, based on unsupervised entity linking. The used KB entity representations are learned from either (i) hyperlinked text documents (Wikipedia), or (ii) a knowledge graph (Wikidata), and appear complementary in raising IE performance. Representations of corresponding entity linking (EL) candidates are added to text span representations of the input document, and we experiment with (i) taking a weighted average of the EL candidate representations based on their prior (in Wikipedia), and (ii) using an attention scheme over the EL candidate list. Results demonstrate an increase of up to 5% F1-score for the evaluated IE tasks on two datasets. Despite a strong performance of the prior-based model, our quantitative and qualitative analysis reveals the advantage of using the attention-based approach.

Injecting Knowledge Base Information into End-to-End Joint Entity and Relation Extraction and Coreference Resolution

Though being a primary trend for enhancing interpretability of neural networks, attention mechanism's reliability and validity are still under debate. In this paper, we try to purify attention scores to obtain a more faithful explanation of downstream models. Specifically, we propose a framework consisting of a learner and a compressor, which performs fine-tuning and compressing iteratively to enhance the performance and interpretability of the attention mechanism. The learner focuses on learning better text representations to achieve good decisions by fine-tuning, while the compressor aims to perform compressions over the representations to retain the most useful clues for explanations with a Variational information bottleneck ATtention (VAT) mechanism. Extensive experiments on eight benchmark datasets show the great advantages of our proposed approach in terms of both performance and interpretability.

Content not yet available

On the Interplay Between Fine-tuning and Composition in Transformers

Next from IJCNLP-AACL 2021

Dynamic Contextualized Word Embeddings

Similar lecture

How Good is Your Tokenizer? On the Monolingual Performance of Multilingual Language Models