Ireland

Recent studies have found that removing the norm-bounded projection and increasing search steps in adversarial training can significantly improve robustness. However, we observe that a too large number of search steps can hurt accuracy. We aim to obtain strong robustness efficiently using fewer steps. Through a toy experiment, we find that perturbing the clean data to the decision boundary but not crossing it does not degrade the test accuracy. Inspired by this, we propose friendly adversarial data augmentation (FADA) to generate friendly adversarial data. On top of FADA, we propose geometry-aware adversarial training (GAT) to perform adversarial training on friendly adversarial data so that we can save a large number
of search steps. Comprehensive experiments across two widely used datasets and three pretrained language models demonstrate that GAT can obtain stronger robustness via fewer steps. In addition, we provide extensive empirical results and in-depth analyses on robustness to facilitate future studies.

ACL 2022

Improving Robustness of Language Models from a Geometry-aware Perspective

natural language processing

adversarial training

adversarial robustness

## Welcome!
"I am excited to welcome you to this year’s edition of the Conference on Empirical Methods in Natural Language Processing! Importantly, it marks the 30th edition of EMNLP. With over 8,000 submissions, more than 3,000 accepted papers, and thousands of attendees, we have come a long way from that first
workshop, which had 14 accepted papers. As the field looks ahead, Suzhou is the fitting location for celebrating this milestone: rooted in a long literary tradition, yet modern and forward-looking, and home to a large share of the NLP community." 

*Message from the General Chair, Dirk Hovy*

EMNLP 2025

Celebrate 30 Years of EMNLP! 
EMNLP 2025 will be held in Suzhou, China from November 5th to November 9th, 2025.

findings / work in progress

# Welcome everyone to ACL 2022!

The 60th Annual Meeting of the Association for Computational Linguistics is taking place May 22-27, 2022 as a hybrid event, in Dublin and online. We are happy to welcome all of you to this anniversary edition with an almost 50-50 in-person and virtual participation. 
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, three amazing keynote events and two new initiatives of invited talks: Spotlight Talks for Young Rising Stars (STIRS) and The Next Big Idea Talks. Posters (including Findings of ACL 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 will have an oral session and a poster session as part of Poster Session 1. The program also features eight Tutorials and 28 Workshops. 

 
We wish you a wonderful conference! 
[**The ACL 2022 Organizing Committee**](https://www.2022.aclweb.org/organisers)
 
[**Conference Handbook**](https://drive.google.com/file/d/1_BUCMfhMVrjG9E2e71aHdHeE28KSje0l/view?usp=sharing) 
[**Mini Handbook**](https://drive.google.com/file/d/1qlBKl0wzmlVF1oCeMQl3BahLd9nLP5Ce/view?usp=sharing) 
[**Posters and Demo guides**](https://drive.google.com/file/d/1UucMAoCNncIOaH1rMMDa0owuG9qgvJTG/view?usp=sharing)

The Association for Computational Linguistics (ACL) is the premier international scientific and professional society for people working on computational problems involving human language, a field often referred to as either computational linguistics or natural language processing (NLP). 

Modern Natural Language Processing (NLP) models are known to be sensitive to input perturbations and their performance can decrease when applied to real-world, noisy data. However, it is still unclear why models are less robust to some perturbations than others. In this work, we test the hypothesis that the extent to which a model is affected by an unseen textual perturbation (robustness) can be explained by the learnability of the perturbation (defined as how well the model learns to identify the perturbation with a small amount of evidence). We further give a causal justification for the learnability metric. We conduct extensive experiments with four prominent NLP models -- TextRNN, BERT, RoBERTa and XLNet -- over eight types of textual perturbations on three datasets. We show that a model which is better at identifying a perturbation (higher learnability) becomes worse at ignoring such a perturbation at test time (lower robustness), providing empirical support for our hypothesis.

Interpreting the Robustness of Neural NLP Models to Textual Perturbations

Traditional methods for named entity recognition (NER) classify mentions into a fixed set of pre-defined entity types. However, in many real-world scenarios, 
new entity types are incrementally involved. To investigate this problem, continual learning is introduced for NER. However, the existing method depends on the relevance between tasks and is prone to inter-type confusion.
In this paper, we propose a novel two-stage framework Learn-and-Review (L\&R) for continual NER under the type-incremental setting to alleviate the above issues.
Specifically, for the learning stage, we distill the old knowledge from teacher to a student on the current dataset. For the reviewing stage, we first generate synthetic samples of old types to augment the dataset. Then, we further distill new knowledge from the above student and old knowledge from the teacher to get an enhanced student on the augmented dataset. This stage has the following advantages: (1) The synthetic samples mitigate the gap between the old and new task and thus enhance the further distillation; (2) Different types of entities are jointly seen during training which alleviates the inter-type confusion. Experimental results show that L\&R outperforms the state-of-the-art method on CoNLL-03 and OntoNotes-5.0.

Learn and Review: Enhancing Continual Named Entity Recognition via Reviewing Synthetic Samples

Language models excel at generating coherent text, and model compression techniques such as knowledge distillation have enabled their use in resource-constrained settings. However, these models can be biased in multiple ways, including the unfounded association of male and female genders with gender-neutral professions. Therefore, knowledge distillation without any fairness constraints may preserve or exaggerate the teacher model’s biases onto the distilled model. To this end, we present a novel approach to mitigate gender disparity in text generation by learning a fair model during knowledge distillation. We propose two modifications to the base knowledge distillation based on counterfactual role reversal—modifying teacher probabilities and augmenting the training set. We evaluate gender polarity across professions in open-ended text generated from the resulting distilled and finetuned GPT–2 models and demonstrate a substantial reduction in gender disparity with only a minor compromise in utility. Finally, we observe that language models that reduce gender polarity in language generation do not improve embedding fairness or downstream classification fairness.

Mitigating Gender Bias in Distilled Language Models via Counterfactual Role Reversal

Transcription is often reported as the bottleneck in endangered language documentation, requiring large efforts from scarce speakers and transcribers. In general, automatic speech recognition (ASR) can be accurate enough to accelerate transcription only if trained on large amounts of transcribed data. However, when a single speaker is involved, several studies have reported encouraging results for phonetic transcription even with small amounts of training. Here we expand this body of work on speaker-dependent transcription by comparing four ASR approaches, notably recent transformer and pretrained multilingual models, on a common dataset of 11 languages. To automate data preparation, training and evaluation steps, we also developed a phoneme recognition setup which handles morphologically complex languages and writing systems for which no pronunciation dictionary exists.
We find that fine-tuning a multilingual pretrained model yields an average phoneme error rate (PER) of 15% for 6 languages with 99 minutes or less of transcribed data for training. For the 5 languages with between 100 and 192 minutes of training, we achieved a PER of 5.9% or less. 
These results on a number of varied languages suggest that ASR can now significantly reduce transcription efforts in the speaker-dependent situation common in endangered language work.

Phoneme transcription of endangered languages: an evaluation of recent ASR architectures in the single speaker scenario

A common method for extractive multi-document news summarization is to re-formulate it as a single-document summarization problem by concatenating all documents as a single meta-document. However, this method neglects the relative importance of documents. We propose a simple approach to reorder the documents according to their relative importance before concatenating and summarizing them. The reordering makes the salient content easier to learn by the summarization model. Experiments show that our approach outperforms previous state-of-the-art methods with more complex architectures.

Read Top News First: A Document Reordering Approach for Multi-Document News Summarization

We present RuCCoN, a new dataset for clinical concept normalization in Russian manually annotated by medical professionals. It contains over 16,028 entity mentions manually linked to over 2,409 unique concepts from the Russian language part of the UMLS ontology. 
We provide train/test splits for different settings (stratified, zero-shot, and CUI-less) and present strong baselines obtained with state-of-the-art models such as SapBERT. At present, Russian medical NLP is lacking in both datasets and trained models, and we view this work as an important step towards filling this gap. Our dataset and annotation guidelines are available at https://github.com/AIRI-Institute/RuCCoN .

RuCCoN: Clinical Concept Normalization in Russian

With the increasing popularity of online chatting, stickers are becoming important in our online communication. Selecting appropriate stickers in open-domain dialogue requires a comprehensive understanding of both dialogues and stickers, as well as the relationship between the two types of modalities. To tackle these challenges, we propose a multitask learning method comprised of three auxiliary tasks to enhance the understanding of dialogue history, emotion and semantic meaning of stickers. Extensive experiments conducted on a recent challenging dataset show that our model can better combine the multimodal information and achieve significantly higher accuracy over strong baselines. Ablation study further verifies the effectiveness of each auxiliary task.

Selecting Stickers in Open-Domain Dialogue through Multitask Learning

Aspect-based sentiment analysis (ABSA) tasks aim to extract sentiment tuples from a sentence. Recent generative methods such as Seq2Seq models have achieved good performance by formulating the output as a sequence of sentiment tuples. However, the orders between the sentiment tuples do not naturally exist and the generation of the current tuple should not condition on the previous ones. In this paper, we propose Seq2Path to generate sentiment tuples as paths of a tree. A tree can represent ``1-to-n'' relations (e.g., an aspect term may correspond to multiple opinion terms) and the paths of a tree are independent and do not have orders. For training, we treat each path as an independent target, and we calculate the average loss of the ordinary Seq2Seq model over paths. For inference, we apply beam search with constrained decoding. By introducing an additional discriminative token and applying a data augmentation technique, valid paths can be automatically selected. We conduct experiments on five tasks including AOPE, ASTE, TASD, UABSA, ACOS. We evaluate our method on four common benchmark datasets including Laptop14, Rest14, Rest15, Rest16. Our proposed method achieves state-of-the-art results in almost all cases.

Seq2Path: Generating Sentiment Tuples as Paths of a Tree

The prototypical NLP experiment trains a standard architecture on labeled English data and optimizes for accuracy, without accounting for other dimensions such as fairness, interpretability, or computational efficiency. We show through a manual classification of recent NLP research papers that this is indeed the case and refer to it as the Square One experimental setup. We observe that NLP research often goes beyond the square one setup, e.g, focusing not only on accuracy, but also on fairness or interpretability, but typically only along a single dimension. Most work targeting multilinguality, for example, considers only accuracy; most work on fairness or interpretability considers only English; and so on.
Such one-dimensionality of most research means we are only exploring a fraction of the NLP research search space. We provide historical and recent examples of how the square one bias has led researchers to draw false conclusions or make unwise choices, point to promising yet unexplored directions on the research manifold, and make practical recommendations to enable more multi-dimensional research. We open-source the results of our annotations to enable further analysis.

Square One Bias in NLP: Towards a Multi-Dimensional Exploration of the Research Manifold

As more and more pre-trained language models adopt on-cloud deployment, the privacy issues grow quickly, mainly for the exposure of plain-text user data (e.g., search history, medical record, bank account). Privacy-preserving inference of transformer models is on the demand of cloud service users. To protect privacy, it is an attractive choice to compute only with ciphertext in homomorphic encryption (HE). However, enabling pre-trained models inference on ciphertext data is difficult due to the complex computations in transformer blocks, which are not supported by current HE tools yet. In this work, the speaker will introduce THE-X, an approximation approach for transformers, which enables privacy-preserving inference of pre-trained models developed by popular frameworks. THE-X can enable transformer inference on encrypted data for different downstream tasks, with negligible performance drop but enjoying the theory-guaranteed privacy-preserving advantage.

Downloads

Next from ACL 2022

Interpreting the Robustness of Neural NLP Models to Textual Perturbations

Similar lecture

Adversarial Defence without Adversarial Defence: Enhancing Language Model Robustness via Instance-level Principal Component Removal