United States

Recent advancements in open-domain question answering (ODQA), that is, finding answers from large open-domain corpus like Wikipedia, have led to human-level performance on many datasets. However, progress in QA over book stories (Book QA) lags despite its similar task formulation to ODQA. This work provides a comprehensive and quantitative analysis about the difficulty of Book QA: (1) We benchmark the research on the NarrativeQA dataset with extensive experiments with cutting-edge ODQA techniques. This quantifies the challenges Book QA poses, as well as advances the published state-of-the-art with a ∼7% absolute improvement on ROUGE-L. (2) We further analyze the detailed challenges in Book QA through human studies.1 Our findings indicate that the event-centric questions dominate this task, which exemplifies the inability of existing QA models to handle event-oriented scenarios.

NAACL 2022

Narrative Question Answering with Cutting-Edge Open-Domain QA Techniques: A Comprehensive Study

## Welcome to NAACL 2022!
The Annual Meeting of the North American Chapter of the Association for Computational Linguistics (NAACL) is taking place July 10-15, 2022 as a hybrid event, in Seattle, WA and online. We are happy to welcome all of you to conference! 
 
The main conference program features oral presentations, in-person and virtual posters and demo sessions, a plenary session for our best paper presentations and awards, keynote presentations, a plenary panel on the place of linguistics and symbolic structures in NLP, and an Industrial Track panel on Careers in NLP.
 
Posters (including Findings of NAACL 2022) and demos are grouped by areas for both the in-person and the virtual sessions. For the virtual component, the talks will be on Zoom and the posters and the demos will be in GatherTown. The Student Research Workshop and will have an oral session and a poster session.
 
The program also features 6 Tutorials and 26 Workshops.
 
 We wish you a wonderful conference!
 
[The NAACL 2022 Organizing Committee](https://2022.naacl.org/committees/organization/)

2022 Annual Conference of the North American Chapter of the Association for Computational Linguistics

poster

Prior works on the open-domain dialogue primarily focused on exploiting self personas. We propose a novel framework that automatically generates partner personas and enhances the succeeding response generation with the partner personas generated with the use of reinforcement learning.

Partner Personas Generation for Dialogue Response Generation

Dense retrieval approaches can overcome the lexical gap and lead to significantly improved search results. However, they require large amounts of training data which is not available for most domains. As shown in previous work (Thakur et al., 2021b), the performance of dense retrievers severely degrades under a domain shift. This limits the usage of dense retrieval approaches to only a few domains with large training datasets.
In this paper, we propose the novel unsupervised domain adaptation method Generative Pseudo Labeling (GPL), which combines a query generator with pseudo labeling from a cross-encoder. On six representative domain-specialized datasets, we find the proposed GPL can outperform an out-of-the-box state-of-the-art dense retrieval approach by up to 9.3 points nDCG@10. GPL requires less (unlabeled) data from the target domain and is more robust in its training than previous methods.
We further investigate the role of six recent pre-training methods in the scenario of domain adaptation for retrieval tasks, where only three could yield improved results. The best approach, TSDAE (Wang et al., 2021) can be combined with GPL, yielding another average improvement of 1.4 points nDCG@10 across the six tasks. The code and the models are available at https://github.com/UKPLab/gpl.

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

We use paraphrases as a unique source of data to analyze contextualized embeddings, with a particular focus on BERT. Because paraphrases naturally encode consistent word and phrase semantics, they provide a unique lens for investigating properties of embeddings. Using the Paraphrase Database's alignments, we study words within paraphrases as well as phrase representations. We find that contextual embeddings effectively handle polysemous words, but give synonyms surprisingly different representations in many cases. We confirm previous findings that BERT is sensitive to word order, but find slightly different patterns than prior work in terms of the level of contextualization across BERT's layers.

Using Paraphrases to Study Properties of Contextual Embeddings

Distributional models learn representations of words from text, but are criticized for their lack of grounding, or the linking of text to the non-linguistic world. Grounded language models have had success in learning to connect concrete categories like nouns and adjectives to the world via images and videos, but can struggle to isolate the meaning of verbs from the context in which they typically occur, e.g. a person walking across a tennis court is labeled “swinging a tennis racket". We are interested in building representations of trajectories (i.e. position and rotation of objects over time), and evaluating the extent to which they encode verb semantics. To achieve this, we build a procedurally generated agent-object-interaction dataset, and obtain human annotations for the verbs that occur. We find that trajectory data correlates as-is with verbs like fall, and can be abstracted through self-supervised pretraining to representations that capture nuanced differences in verb meaning, such as roll vs. slide, while falling short on other distinctions like fall vs. drop.

Do Trajectories Encode Verb Meaning?

Czech Grammar Error Correction with a Large and Diverse Corpus

We establish THumB, a rubric-based human evaluation protocol for image captioning models. Our scoring rubrics and their definitions are carefully developed based on machine- and human-generated captions on the MSCOCO dataset. Each caption is evaluated along two main dimensions in a tradeoff (precision and recall) as well as other aspects that measure the text quality (fluency, conciseness, and inclusive language). Our evaluations demonstrate several critical problems of the current evaluation practice. Human-generated captions show substantially higher quality than machine-generated ones, especially in coverage of salient information (i.e., recall), while most automatic metrics say the opposite. Our rubric-based results reveal that CLIPScore, a recent metric that uses image features, better correlates with human judgments than conventional text-only metrics because it is more sensitive to recall. We hope that this work will promote a more transparent evaluation protocol for image captioning and its automatic metrics.

Transparent Human Evaluation for Image Captioning

The Yes/No QA task (Clark et al., 2019) consists of "Yes" or "No" questions about a given context. However, in realistic scenarios, the information provided in the context is not always sufficient in order to answer the question. For example,  given the context "She married a lawyer from New-York.", we don't know whether the answer to the question "Did she marry in New York?" is "Yes" or "No". In this paper, we extend the Yes/No QA task, adding questions with an IDK answer, and show its considerable difficulty compared to the original 2-label task. For this purpose, we (i) enrich the BoolQ dataset (Clark et al., 2019) to include unanswerable questions and (ii) create out-of-domain test sets for the Yes/No/IDK QA task. We study the contribution of training on other Natural Language Understanding tasks. We focus in particular on Extractive QA (Rajpurkar et al., 2018) and Recognizing Textual Entailments (RTE; Dagan et al., 2013), analyzing the differences between 2 and 3 labels using the new data.

Yes, No or IDK: The Challenge of Unanswerable Yes/No Questions

Pre-trained Language Models (PTLMs) have been shown to perform well on natural language tasks. Many prior works have leveraged structured commonsense present in the form of entities linked through labeled relations in Knowledge Graphs (KGs) to assist PTLMs. Retrieval approaches use KG as a separate static module which limits coverage since KGs contain finite knowledge. Generative methods train PTLMs on KG triples to improve the scale at which knowledge can be obtained. However, training on symbolic KG entities limits their applicability in tasks involving natural language text where they ignore overall context. To mitigate this, we propose a CommonSense Contextualizer (CoSe-Co) conditioned on sentences as input to make it generically usable in tasks for generating knowledge relevant to the overall context of input text. To train CoSe-Co, we propose a novel dataset comprising of sentence and commonsense knowledge pairs. The knowledge inferred by CoSe-Co is diverse and contain novel entities not present in the underlying KG. We augment generated knowledge in Multi-Choice QA and Open-ended CommonSense Reasoning tasks leading to improvements over current best methods on CSQA, ARC, QASC and OBQA datasets. We also demonstrate its applicability in improving performance of a baseline model for paraphrase generation task.

CoSe-Co: Text Conditioned Generative CommonSense Contextualizer

Global models are typically trained to be as generalizable as possible. Invariance to the specific user is considered desirable since models are shared across multitudes of users. However, these models are often unable to produce personalized responses for individual users, based on their data.  Contrary to widely-used personalization techniques based on few-shot and meta-learning, we propose UserIdentifier, a novel scheme for training a single shared model for all users. Our approach produces personalized responses by prepending a fixed, user-specific non-trainable string (called ``user identifier'') to each user's input text. Unlike prior work, this method doesn't need any additional model parameters, any extra rounds of personal few-shot learning or any change made to the vocabulary. We empirically study different types of user identifiers (numeric, alphanumeric, and also randomly generated) and demonstrate that, surprisingly, randomly generated user identifiers outperform the prefix-tuning based state-of-the-art approach by up to 13, on a suite of sentiment analysis datasets. 

UserIdentifier: Implicit User Representations for Simple and Effective Personalized Sentiment Analysis

Recognizing offensive text is an important requirement for every content management system, especially for social networks. While the majority of the prior work formulate this problem as text classification, i.e., if a text excerpt is offensive or not, in this work we propose a novel model for offensive span detection (OSD), whose goal is to identify the spans responsible for the offensive tone of the text. One of the challenges to train a model for this novel setting is the lack of enough training data. To address this limitation, in this work we propose a novel method in which the large-scale pre-trained language model GPT-2 is employed to generate synthetic training data for OSD. In particular, we propose to train the GPT-2 model in a dual-training setting using the REINFORCE algorithm to generate in-domain, natural and diverse training samples. Extensive experiments on the benchmark dataset for OSD reveal the effectiveness of the proposed method.

Narrative Question Answering with Cutting-Edge Open-Domain QA Techniques: A Comprehensive Study

Downloads

Next from NAACL 2022

Partner Personas Generation for Dialogue Response Generation

Similar lecture

Seeing the wood for the trees: a contrastive regularization method for the low-resource Knowledge Base Question Answering

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES