United Arab Emirates

General pre-trained language models (PLMs), such as BERT, have achieved remarkable performance on various NLP tasks. Recently, domain-specific PLMs have been proposed to boost the task performance of specific domains (e.g., biomedical and computer science) by continuing to pre-train general PLMs with domain-specific corpora. However, this domain-adaptive pre-training (DAPT \cite{DBLP:conf/acl/GururanganMSLBD20}) tends to forget the previous general knowledge acquired by general PLMs, which leads to a \emph{catastrophic forgetting} phenomenon and sub-optimal performance. To alleviate this problem, we propose a new framework of \textbf{M}emory-\textbf{A}ugmented \textbf{P}re-trained Language Model (\textbf{MAP}), which augments the domain-specific PLM by a memory built from the frozen general PLM without losing the general knowledge. Specifically, we propose a new memory-augmented layer, and based on it, different augmentation strategies are explored to build memory and fusion memory into domain-specific PLM. We demonstrate the effectiveness of MAP on different domains (biomedical and computer science publications, news, and reviews) and different kinds (text classification, QA, NER) of tasks, and the extensive results show that the proposed MAP can achieve SOTA results on these tasks.

EMNLP 2022

G-MAP: General Memory-Augmented Pre-trained Language Model for Domain Tasks

domain-adaptive pre-training

memory-augmented

catastrophic forgetting

## 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/)

NAACL 2022

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

poster

## Welcome to EMNLP 2022!
I am delighted to welcome you to EMNLP 2022! I believe this conference has been complicated beyond any precedent. Over the past year, it’s been thrilling to see the organization team approach each new puzzle with creativity and enthusiasm. We hope that those participating in Abu Dhabi as well as those joining remotely will leave the conference feeling newly inspired by the program and newly connected to our ever-growing community. Following EMNLP 2021 and major NLP conferences since, EMNLP 2022 is “hybrid,” serving both virtual and in-person participants.

Our key innovations for EMNLP 2022 include:

* EMNLP 2022 is “hybrid” in a second sense, as well: we allowed both direct and rolling review paper submissions, building on the pilot experiment of EMNLP 2021, which considered a small number of ARR submissions. 
* Familiar from NAACL but new to EMNLP, we’ve added an industry track.
* During the conference, “portals” will link virtual poster sessions to in-person conference participants during poster sessions each day.
* The first ACL-family conference in the United Arab Emirates.

 *Message from Noah A. Smith, University of Washington and Allen Institute for AI, Seattle, Washington, USA* 
***EMNLP 2022 General Chair***
 
[![](https://assets.underline.io/uploads/markdown_image/1/image/9eec7d4a287ee18c278b08229290aa83.png)](https://drive.google.com/file/d/1OlPv6QBeo62VVTughj2jkiLeyHd1WnUt/view)
 
[![](https://assets.underline.io/uploads/markdown_image/1/image/a3db7a768409f05192210d98601edb25.png)](https://emnlp2022.rocket.chat/)

To access this site you need to register. Please register [here](https://2022.emnlp.org/registration/).

Register here

Welcome!
EMNLP 2022 will take place in Abu Dhabi from December 7th to December 11th, 2022. And it will be held in hybrid mode, both online and offline.

Despite the recent success of large pretrained language models in NLP, they are susceptible to adversarial examples. Concurrently, several studies on adversarial images have observed an intriguing property: the adversarial images tend to leave the low-dimensional natural data manifold. In this study, we find a similar phenomenon occurs in the contextualized embedding space of natural sentences induced by pretrained language models in which textual adversarial examples tend to have their embeddings diverge off the manifold of natural sentence embeddings. Based on this finding, we propose Textual Manifold-based Defense (TMD), a defense mechanism that learns the embedding space manifold of the underlying language model and projects novel inputs back to the approximated structure before classification. Through extensive experiments, we find that our method consistently and significantly outperforms previous defenses under various attack settings while remaining unaffected to the clean accuracy. To the best of our knowledge, this is the first kind of manifold-based defense adapted to the NLP domain.

Textual Manifold-based Defense Against Natural Language Adversarial Examples

Task-oriented dialogue (TOD) systems have been widely used by mobile phone intelligent assistants to accomplish tasks such as calendar scheduling or hotel reservation. Current TOD systems usually focus on multi-turn text/speech interaction, then they would call back-end APIs designed for TODs to perform the task. However, this API-based architecture greatly limits the information-searching capability of intelligent assistants and may even lead to task failure if TOD-specific APIs are not available or the task is too complicated to be executed by the provided APIs. In this paper, we propose a new TOD architecture: GUI-based task-oriented dialogue system (GUI-TOD). A GUI-TOD system can directly perform GUI operations on real APPs and execute tasks without invoking TOD-specific backend APIs. Furthermore, we release META-GUI, a dataset for training a Multi-modal convErsaTional Agent on mobile GUI. We also propose a multi-model action prediction and response model, which show promising results on META-GUI. The dataset, codes and leaderboard are publicly available.

META-GUI: Towards Multi-modal Conversational Agents on Mobile GUI

Recent years have witnessed the prevalent application of pre-trained language models (PLMs) in NLP. From the perspective of parameter space, PLMs provide generic initialization, starting from which high-performance minima could be found. Although plenty of works have studied how to effectively and efficiently adapt PLMs to high-performance minima, little is known about the connection of various minima reached under different adaptation configurations. In this paper, we investigate the geometric connections of different minima through the lens of mode connectivity, which measures whether two minima can be connected with a low-loss path. We conduct empirical analyses to investigate three questions: (1) how could hyperparameters, specific tuning methods, and training data affect PLM's mode connectivity? (2) How does mode connectivity change during pre-training? (3) How does the PLM's task knowledge change along the path connecting two minima? In general, exploring the mode connectivity of PLMs conduces to understanding the geometric connection of different minima, which may help us fathom the inner workings of PLM downstream adaptation. The codes are publicly available at https://github.com/thunlp/Mode-Connectivity-PLM.

Exploring Mode Connectivity for Pre-trained Language Models

Prompt Tuning has been largely successful as a parameter-efficient method of conditioning large-scale pre-trained language models to perform downstream tasks. Thus far, soft prompt tuning learns a fixed set of task-specific continuous vectors, i.e., soft tokens that remain static across the task samples. A fixed prompt, however, may not generalize well to the diverse kinds of inputs the task comprises. In order to address this, we propose Vector-quantized Input-contextualized Prompts (VIP) as an extension to the soft prompt tuning framework. VIP particularly focuses on two aspects---contextual prompts that learns input-specific contextualization of the soft prompt tokens through a small-scale sentence encoder and quantized prompts that maps the contextualized prompts to a set of learnable codebook vectors through a Vector quantization network. On various language understanding tasks like SuperGLUE, QA, Relation classification, NER and NLI, VIP outperforms the soft prompt tuning (PT) baseline by an average margin of 1.19%. Further, our generalization studies show that VIP learns more robust prompt representations, surpassing PT by a margin of 0.6% - 5.3% on Out-of-domain QA and NLI tasks respectively, and by 0.75% on Multi-Task setup over 4 tasks spanning across 12 domains.

Vector-Quantized Input-Contextualized Soft Prompts for Natural Language Understanding

Pre-training language models have achieved thriving success in numerous natural language understanding and autoregressive generation tasks, but non-autoregressive generation in applications such as machine translation has not sufficiently benefited from the pre-training paradigm. In this work, we establish the connection between a pre-trained masked language model (MLM) and non-autoregressive generation on machine translation. From this perspective, we present XLM-D, which seamlessly transforms an off-the-shelf cross-lingual pre-training model into a non-autoregressive translation (NAT) model with a lightweight yet effective decorator. Specifically, the decorator ensures the representation consistency of the pre-trained model and brings only one additional trainable parameter. Extensive experiments on typical translation datasets show that our models obtain state-of-the-art performance while realizing the inference speed-up by 19.9x. One striking result is that on WMT14 En-De, our XLM-D obtains 29.80 BLEU points with multiple iterations, which outperforms the previous mask-predict model by 2.77 points.

XLM-D: Decorate Cross-lingual Pre-training Model as Non-Autoregressive Neural Machine Translation

To address the Conversational Question Answering (ORConvQA) task, previous work has considered an effective three-stage architecture, consisting of a retriever, a reranker, and a reader to extract the answers. In order to effectively answer the users' questions, a number of existing approaches have applied multi-task learning, such that the same model is shared between the reranker and the reader. Such approaches also typically tackle reranking and reading as classification tasks. On the other hand, recent text generation models, such as monoT5 and UnifiedQA, have been shown to respectively yield impressive performances in passage reranking and reading. However, no prior work has combined monoT5 and UnifiedQA to share a single text generation model that directly extracts the answers for the users instead of predicting the start/end positions in a retrieved passage. In this paper, we investigate the use of Multi-Task Learning (MTL) to improve performance on the ORConvQA task by sharing the reranker and reader's learned structure in a generative model. In particular, we propose monoQA, which uses a text generation model with multi-task learning for both the reranker and reader. Our model, which is based on the T5 text generation model, is fine-tuned simultaneously for both reranking (in order to improve the precision of the top retrieved passages) and extracting the answer. Our results on the OR-QuAC and OR-CoQA datasets demonstrate the effectiveness of our proposed model, which significantly outperforms existing strong baselines with improvements ranging from +12.31% to +19.51% in MAP and from +5.70% to +23.34% in F1 on all used test sets.

monoQA: Multi-Task Learning of Reranking and Answer Extraction for Open-Retrieval Conversational Question Answering

Keyphrase generation aims to automatically generate short phrases summarizing an input document. The recently emerged ONE2SET paradigm (Ye et al., 2021) generates keyphrases as a set and has achieved competitive performance. Nevertheless, we observe serious calibration errors outputted by ONE2SET, especially in the over-estimation of ∅ token (means "no corresponding keyphrase”). In this paper, we deeply analyze this limitation and identify two main reasons behind: 1) the parallel generation has to introduce excessive ∅ as padding tokens into training instances; and 2) the training mechanism assigning target to each slot is unstable and further aggravates the ∅ token over-estimation. To make the model well-calibrated, we propose WR-ONE2SET which extends ONE2SET with an adaptive instance-level cost Weighting strategy and a target Re-assignment mechanism. The former dynamically penalizes the over-estimated slots for different instances thus smoothing the uneven training distribution. The latter refines the original inappropriate assignment and reduces the supervisory signals of over-estimated slots. Experimental results on commonly-used datasets demonstrate the effectiveness and generality of our proposed paradigm.

WR-One2Set: Towards Well-Calibrated Keyphrase Generation

Data scarcity is a widespread problem for numerous natural language processing (NLP) tasks within low-resource languages. Within morphology, the labour-intensive task of tagging/glossing data is a serious bottleneck for both NLP and fieldwork. Active learning (AL) aims to reduce the cost of data annotation by selecting data that is most informative for the model. In this paper, we explore four sampling strategies for the task of morphological inflection using a Transformer model: a pair of oracle experiments where data is chosen based on correct/incorrect predictions by the model, model confidence, entropy, and random selection. We investigate the robustness of each sampling strategy across 30 typologically diverse languages, as well as a 10-cycle iteration using Natügu as a case study. Our results show a clear benefit to selecting data based on model confidence. Unsurprisingly, the oracle experiment, which is presented as a proxy for linguist/language informer feedback, shows the most improvement. This is followed closely by low-confidence and high-entropy forms. We also show that despite the conventional wisdom of larger data sets yielding better accuracy, introducing more instances of high-confidence, low-entropy, or forms that the model can already inflect correctly, can reduce model performance.

Eeny, meeny, miny, moe. How to choose data for morphological inflection.

Detecting out-of-domain (OOD) intents from user queries is essential for avoiding wrong operations in task-oriented dialogue systems. The key challenge is how to distinguish in-domain (IND) and OOD intents. Previous methods ignore the alignment between representation learning and scoring function, limiting the OOD detection performance. In this paper, we propose a unified neighborhood learning framework (UniNL) to detect OOD intents. Specifically, we design a KNCL objective for representation learning, and introduce a KNN-based scoring function for OOD detection. We aim to align representation learning with scoring function. Experiments and analysis on two benchmark datasets show the effectiveness of our method.

UniNL: Aligning Representation Learning with Scoring Function for OOD Detection via Unified Neighborhood Learning

Preserving privacy in contemporary NLP models allows us to work with sensitive data, but unfortunately comes at a price. We know that stricter privacy guarantees in differentially-private stochastic gradient descent (DP-SGD) generally degrade model performance. However, previous research on the efficiency of DP-SGD in NLP is inconclusive or even counter-intuitive. In this short paper, we provide an extensive analysis of different privacy preserving strategies on seven downstream datasets in five different `typical' NLP tasks with varying complexity using modern neural models based on BERT and XtremeDistil architectures. We show that unlike standard non-private approaches to solving NLP tasks, where bigger is usually better, privacy-preserving strategies do not exhibit a winning pattern, and each task and privacy regime requires a special treatment to achieve adequate performance.

Downloads

Next from EMNLP 2022

Textual Manifold-based Defense Against Natural Language Adversarial Examples

Similar lecture

KALA: Knowledge-Augmented Language Model Adaptation

.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 EMNLP 2022

Textual Manifold-based Defense Against Natural Language Adversarial Examples

Similar lecture

KALA: Knowledge-Augmented Language Model Adaptation

Downloads