Thailand

Despite the prominence of neural abstractive summarization models, we know little about how they actually form summaries and how to understand where their decisions come from.
We propose a two-step method to interpret summarization model decisions. 
We first analyze the model&#39;s behavior by ablating the full model to categorize each decoder decision into one of several generation modes: roughly, is the model behaving like a language model, is it relying heavily on the input, or is it somewhere in between?
After isolating decisions that do depend on the input, we explore interpreting these decisions using several different attribution methods.
We compare these techniques based on their ability to select content and reconstruct the model&#39;s predicted token from perturbations of the input, thus revealing whether highlighted attributions are truly important for the generation of the next token.
While this machinery can be broadly useful even beyond summarization, we specifically demonstrate its capability to identify phrases the summarization model has memorized and determine where in the training pipeline this memorization happened, as well as study complex generation phenomena like sentence fusion on a per-instance basis.

IJCNLP-AACL 2021

Dissecting Generation Modes for Abstractive Summarization Models via Ablation and Attribution

model interpretability

text summarization

generation

**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).

Despite the prominence of neural abstractive summarization models, we know little about how they actually form summaries and how to understand where their decisions come from.
We propose a two-step method to interpret summarization model decisions. 
We first analyze the model's behavior by ablating the full model to categorize each decoder decision into one of several generation modes: roughly, is the model behaving like a language model, is it relying heavily on the input, or is it somewhere in between?
After isolating decisions that do depend on the input, we explore interpreting these decisions using several different attribution methods.
We compare these techniques based on their ability to select content and reconstruct the model's predicted token from perturbations of the input, thus revealing whether highlighted attributions are truly important for the generation of the next token.
While this machinery can be broadly useful even beyond summarization, we specifically demonstrate its capability to identify phrases the summarization model has memorized and determine where in the training pipeline this memorization happened, as well as study complex generation phenomena like sentence fusion on a per-instance basis.

technical paper

The Shuffle Test is the most common task to evaluate whether NLP models can measure coherence in text.
Most recent work uses direct supervision on the task; we show that by simply finetuning a RoBERTa model, we can achieve a near perfect accuracy of 97.8%, a state-of-the-art.
We argue that this outstanding performance is unlikely to lead to a good model of text coherence, and suggest that the Shuffle Test should be approached in a Zero-Shot setting: models should be evaluated without being trained on the task itself. We evaluate common models in this setting, such as Generative and Bi-directional Transformers, and find that larger architectures achieve high-performance out-of-the-box. Finally, we suggest the k-Block Shuffle Test, a modification of the original by increasing the size of blocks shuffled. Even though human reader performance remains high (around 95% accuracy), model performance drops from 94% to 78% as block size increases, creating a conceptually simple challenge to benchmark NLP models.

Can Transformer Models Measure Coherence In Text: Re-Thinking the Shuffle Test

While online conversations can cover a vast amount of information in many different formats, abstractive text summarization has primarily focused on modeling solely news articles. This research gap is due, in part, to the lack of standardized datasets for summarizing online discussions. To address this gap, we design annotation protocols motivated by an issues--viewpoints--assertions framework to crowdsource four new datasets on diverse online conversation forms of news comments, discussion forums, community question answering forums, and email threads. We benchmark state-of-the-art models on our datasets and analyze characteristics associated with the data. To create a comprehensive benchmark, we also evaluate these models on widely-used conversation summarization datasets to establish strong baselines in this domain. Furthermore, we incorporate argument mining through graph construction to directly model the issues, viewpoints, and assertions present in a conversation and filter noisy input, showing comparable or improved results according to automatic and human evaluations.

ConvoSumm: Conversation Summarization Benchmark and Improved Abstractive Summarization with Argument Mining

Recent years have brought about an interest in the challenging task of summarizing conversation threads (meetings, online discussions, etc.). Such summaries help analysis of the long text to quickly catch up with the decisions made and thus improve our work or communication efficiency. To spur research in thread summarization, we have developed an abstractive Email Thread Summarization (EmailSum) dataset, which contains human-annotated short (<30 words) and long (<100 words) summaries of 2,549 email threads (each containing 3 to 10 emails) over a wide variety of topics. We perform a comprehensive empirical study to explore different summarization techniques (including extractive and abstractive methods, single-document and hierarchical models, as well as transfer and semisupervised learning) and conduct human evaluations on both short and long summary generation tasks. Our results reveal the key challenges of current abstractive summarization models in this task, such as understanding the sender’s intent and identifying the roles of sender and receiver. Furthermore, we find that widely used automatic evaluation metrics (ROUGE, BERTScore) are weakly correlated with human judgments on this email thread summarization task. Hence, we emphasize the importance of human evaluation and the development of better metrics by the community.

EmailSum: Abstractive Email Thread Summarization

Parallel cross-lingual summarization data is scarce, requiring models to better use the limited available cross-lingual resources. Existing methods to do so often adopt sequence-to-sequence networks with multi-task frameworks. Such approaches apply multiple decoders, each of which is utilized for a specific task. However, these independent decoders share no parameters, hence fail to capture the relationships between the discrete phrases of summaries in different languages, breaking the connections in order to transfer the knowledge of the high-resource languages to low-resource languages. To bridge these connections, we propose a novel Multi-Task framework for Cross-Lingual Abstractive Summarization (MCLAS) in a low-resource setting. Employing one unified decoder to generate the sequential concatenation of monolingual and cross-lingual summaries, MCLAS makes the monolingual summarization task a prerequisite of the CLS task. In this way, the shared decoder learns interactions involving alignments and summary patterns across languages, which encourages attaining knowledge transfer. Experiments on two CLS datasets demonstrate that our model significantly outperforms three baseline models in both low-resource and full-dataset scenarios. Moreover, in-depth analysis on the generated summaries and attention heads verifies that interactions are learned well using MCLAS, which benefits the CLS task under limited parallel resources.

Cross-Lingual Abstractive Summarization with Limited Parallel Resources

Recent advances in natural language processing (NLP) have the ability to transform how classroom learning takes place. Combined with the increasing integration of technology in today's classrooms, NLP systems leveraging question answering and dialog processing techniques can serve as private tutors or participants in classroom discussions to increase student engagement and learning. 
To progress towards this goal, we use the classroom discourse framework of academically productive talk (APT) to learn strategies that make for the best learning experience. In this paper, we introduce a new task, called future talk move prediction (FTMP): it consists of predicting the next talk move -- an utterance strategy from APT -- given a conversation history with its corresponding talk moves. We further introduce a neural network model for this task, which outperforms multiple baselines by a large margin. Finally, we compare our model's performance on FTMP to human performance and show several similarities between the two.

What Would a Teacher Do? {P}redicting Future Talk Moves

Capturing the salient information from an input article has been a long-standing challenge for summarization.
On Wikipedia, most of the wiki pages about people contain a factual table that lists the basic properties of the people.
Illuminatingly, a factual table can be regarded as a natural summary of the key information in the corresponding article.
Thus, in this paper we propose the task of table-guided abstractive biography summarization, which utilizes factual tables to capture important information and then generate a summary of a biography.
We first introduce the TaGS (\underline{Ta}ble-\underline{G}uided \underline{S}ummarization) dataset\footnote{Dataset will be released when this paper published.}, the first large-scale biography summarization dataset with tables.
Next, we report some statistics about this dataset to validate the quality of the dataset.
We also benchmark several commonly used summarization methods on TaGS and hope this will inspire more exciting methods.

BioGen: Generating Biography Summary under Table Guidance on Wikipedia

Simultaneous machine translation (SiMT) involves translating source utterances to the target language in real-time before the speaker utterance is completed. This paper proposes the "multilingual" approach to \simt, where a single model simultaneously translates between multiple language-pairs. This not only results in more efficiency in terms of the number of models and parameters (hence simpler deployment), but may also lead to higher performing models by capturing commonalities among the languages. We further explore simple and effective multilingual architectures based on two strong recently proposed SiMT models. Our results on translating from two Germanic languages (German, Dutch) and three Romance languages (French, Italian, Romanian) into English show (i) the single multilingual model is on-par or better than individual models, and (ii) multilingual SiMT models trained based on language families are on-par or better than the universal model trained for all languages.

Multilingual Simultaneous Neural Machine Translation

Supervised learning methods have proven to be effective for Aspect-Based Sentiment Analysis (ABSA).
However, the lack of fine-grained labeled data hinders their effectiveness in many domains. To address this issue, unsupervised domain adaptation methods are desired to transfer knowledge from a labeled source domain to any unlabeled target domain. In this paper, we propose a new domain adaptation paradigm called cross-domain review generation (CDRG), which aims to generate target-domain reviews with fine-grained annotation based on the source-domain labeled reviews. To achieve this goal, we propose a two-step approach as a concrete realization of CDRG. It first converts a source-domain review to a domain-independent review by masking its source-specific attributes, and then converts the domain-independent review to a target-domain review with a masked language model pre-trained in the target domain. We further propose two ways to leverage the generated target-domain reviews for two cross-domain ABSA tasks. Extensive experiments demonstrate the superiority of our CDRG-based approaches over the state-of-the-art domain adaptation methods.

Cross-Domain Review Generation for Aspect-Based Sentiment Analysis

We present a novel method for relation extraction (RE) from a single sentence, mapping the sentence and two given entities to a canonical fact in a knowledge graph (KG). Especially in this presumed sentential RE setting, the context of a single sentence is often sparse. This paper introduces the KGPool method to address this sparsity, dynamically expanding the context with additional facts from the KG. It learns the representation of these facts (entity alias, entity descriptions, etc.) using neural methods, supplementing the sentential context. Unlike existing methods that statically use all expanded facts, KGPool conditions this expansion on the sentence. We study the efficacy of KGPool by evaluating it with different neural models and KGs (Wikidata and NYT Freebase). Our experimental evaluation on standard datasets shows that by feeding the KGPool representation into a Graph Neural Network, the overall method is significantly more accurate than state-of-the-art methods.

KGPool: Dynamic Knowledge Graph Context Selection for Relation Extraction

In multi-hop QA, answering complex questions entails iterative document retrieval for finding the missing entity of the question. The main steps of this process are sub-question detection, document retrieval for the sub-question, and generation of a new query for the final document retrieval. However, building a dataset that contains complex questions with sub-questions and their corresponding documents requires costly human annotation. To address the issue, we propose a new method for weakly supervised multi-hop retriever pre-training without human efforts.
Our method includes 1) a pre-training task for generating vector representations of complex questions, 2) a scalable data generation method that produces the nested structure of question and sub-question as weak supervision for pre-training, and 3) a pre-training model structure based on dense encoders. We conduct experiments to compare the performance of our pre-trained retriever with several state-of-the-art models on end-to-end multi-hop QA as well as document retrieval. The experimental results show that our pre-trained retriever is effective and also robust on limited data and computational resources.

Dissecting Generation Modes for Abstractive Summarization Models via Ablation and Attribution

Downloads

Next from IJCNLP-AACL 2021

Can Transformer Models Measure Coherence In Text: Re-Thinking the Shuffle Test

Similar lecture

BoB: BERT Over BERT for Training Persona-based Dialogue Models from Limited Personalized Data

Dissecting Generation Modes for Abstractive Summarization Models via Ablation and Attribution

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from IJCNLP-AACL 2021

Can Transformer Models Measure Coherence In Text: Re-Thinking the Shuffle Test

Similar lecture

BoB: BERT Over BERT for Training Persona-based Dialogue Models from Limited Personalized Data

Downloads