Indonesia

Evaluation of QA systems is very challenging and expensive, with the most reliable approach being human annotations of correctness of answers for questions. Recent works (AVA, BEM) have shown that transformer LM encoder based similarity metrics transfer well for QA evaluation, but they are limited by the usage of a single correct reference answer. We propose a new evaluation metric: SQuArE (Sentence-level QUestion AnsweRing Evaluation), using multiple reference answers (combining multiple correct and incorrect references) for sentence-form QA. We evaluate SQuArE on both sentence-level extractive (Answer Selection) and generative (GenQA) QA systems, across multiple academic and industrial datasets, and show that it outperforms previous baselines and obtains the highest correlation with human annotations.

IJCNLP-AACL 2023

SQUARE: Automatic Question Answering Evaluation using Multiple Positive and Negative References

technical paper

### Welcome to IJCNLP-AACL 2023!
I am honored and happy to welcome you to IJCNLP-AACL 2023! The conference hosts both the 13th
International Joint Conference on Natural Language Processing (IJCNLP) and the 3rd Conference of
the Asia-Pacific Chapter of the Association for Computational Linguistics (AACL). 
The joint conferences of IJCNLP and AACL have been organized by alternating leadership in the
Asia-Pacific region, in odd years by Asian Federation of Natural Language Processing (AFNLP) and
in even years by AACL. The conference is organized by AFNLP this year. 

Considering the current situation that we are not yet fully over the coronavirus pandemic, the
conference is offered in a hybrid mode, off-line in Nusa Dua, Bali, Indonesia, and on-line with the help
of Underline, during November 1-4, 2023. For those attending off-line, I hope that you will find time to
explore the exciting local area, which remains quite warm and pleasant at this time of the year. For
those attending on-line, I hope that the attendance is not much hampered, sometimes even enhanced
and comfortable, though the on-line arrangements have been made without ample time. 

Message from Jong C. Park, Korea Advanced Institute of Science and Technology (KAIST) 
IJCNLP-AACL 2023 General Chair 
###### [Conference Handbook](https://drive.google.com/file/d/1S2JSXa105xJxIkNBfhYmGH20Om6wARBx/view?usp=sharing)

In order to acces this event page, you need to register [here](http://www.ijcnlp-aacl2023.org/)

The 13th International Joint Conference on Natural Language Processing and the 3rd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics

Despite representing roughly a fifth of the world population, African languages are underrepresented in NLP research, in part due to a lack of datasets. While there are individual language-specific datasets for several tasks, only a handful of tasks (e.g. named entity recognition and machine translation) have datasets covering geographical and typologically-diverse African languages.
In this paper, we develop MasakhaNEWS---the largest dataset for news topic classification covering 16 languages widely spoken in Africa. We provide and evaluate a set of baseline models by training classical machine learning models and fine-tuning several language models. Furthermore, we explore several alternatives to full fine-tuning of language models that are better suited for zero-shot and few-shot learning, such as: cross-lingual parameter-efficient fine-tuning (MAD-X), pattern exploiting training (PET), prompting language models (ChatGPT), and prompt-free sentence transformer fine-tuning (SetFit and the co:here embedding API). Our evaluation in few-shot setting, shows that with as little as 10 examples per label, we achieve more than 90% (i.e. 86.0 F1 points) of the performance of full supervised training (92.6 F1 points) leveraging the PET approach. Our work shows that existing supervised approaches work well for all African languages and that language models with only a few supervised samples can reach competitive performance, both findings which demonstrate the applicability of existing NLP techniques for African languages.

MasakhaNEWS: News Topic Classification for African languages

Zero-shot cross-lingual transfer is a central task in multilingual NLP, allowing models trained in languages with more sufficient training resources to generalize to other low-resource languages.
Earlier efforts on this task use parallel corpora, bilingual dictionaries, or other annotated alignment data to improve cross-lingual transferability, which are typically expensive to obtain.
In this paper, we propose a simple yet effective method, SALT, to improve the zero-shot cross-lingual transfer of the multilingual PLM without the help of such external data.
By incorporating code-switching and embedding mixup with self-augmentation,
SALT effectively distills cross-lingual knowledge from the multilingual PLM and enhances its transferability on downstream tasks.
Experimental results on XNLI and PAWS-X show that our method is able to improve zero-shot cross-lingual transferability without external data.

Self-Augmentation Improves Zero-Shot Cross-Lingual Transfer

Transfer learning has led to large gains in performance for nearly all NLP tasks while making downstream models easier and faster to train. This has also been extended to low-resourced languages, with some success. We investigate the properties of cross-lingual transfer learning between ten low-resourced languages, from the perspective of a named entity recognition task. We specifically investigate how much adaptive fine-tuning and the choice of transfer language affect zero-shot transfer performance. We find that models that perform well on a single language often do so at the expense of generalising to others, while models with the best generalisation to other languages suffer in individual language performance. Furthermore, the amount of data overlap between the source and target datasets is a better predictor of transfer performance than either the geographical or genetic distance between the languages.

Analysing Cross-Lingual Transfer in Low-Resourced African Named Entity Recognition

We present an approach for assessing how multilingual large language models (LLMs) learn syntax in terms of multi-formalism syntactic structures. We aim to recover constituent and dependency structures by casting parsing as sequence labeling. To do so, we select a few LLMs and study them on 13 diverse UD treebanks for dependency parsing and 10 treebanks for constituent parsing. Our results show that the framework is consistent across encodings, and notes the role played by factors such as the syntactic formalism, the LLM differences, and the pretraining and assessment data.

Assessment of Pre-Trained Models Across Languages and Grammars

The scarcity of comprehensive, high-quality Question-Answering (QA) datasets in low-resource languages has greatly limited the progress of research on QA for these languages. This has inspired research on Question-Answer Generation (QAG) which seeks to synthetically generate QA pairs and minimize the human effort required to compile labeled datasets. In this paper, we present the first QAG pipeline for the Bengali language, which consists of an answer span extraction model, a question generation model, and roundtrip consistency filtering to discard inconsistent QA pairs. To train our QAG pipeline, we translate SQuAD1.1 and SQuAD2.0 using the state-of-the-art NLLB machine translation model and accurately mark the answer spans using a novel embedding-based answer alignment algorithm to construct two Bengali QA datasets that we show are superior to the only two existing machine-translated datasets in terms of quality and quantity. We use our QAG pipeline to generate more than 170,000 QA pairs to build BanglaQA, a synthetic QA dataset from 16,000 Bengali news articles spanning 5 different news categories. We demonstrate the quality of BanglaQA by human evaluation on a variety of metrics. The best-performing model among several baselines on our dataset achieves an F1 score of 86.14 falling behind human performance of 95.72 F1. Our codebase and curated datasets are publicly available at https://github.com/shihabshahriar16/BengaliQAG.git.

Question Answer Generation in Bengali: Mitigating the scarcity of QA datasets in a low-resource language

With a rise in false, inaccurate, and misleading information in propaganda, news, and social media, real-world Question Answering (QA) systems face the challenges of synthesizing and reasoning over misinformation-polluted contexts to derive correct answers. This urgency gives rise to the need to make QA systems robust to misinformation, a topic previously unexplored. We study the risk of misinformation to QA models by investigating the sensitivity of open-domain QA models to corpus pollution with misinformation documents. We curate both human-written and model-generated false documents that we inject into the evidence corpus of QA models, and assess the impact on the performance of these systems. Experiments show that QA models are vulnerable to even small amounts of evidence contamination brought by misinformation, with large absolute performance drops on all models. Misinformation attack brings more threat when fake documents are produced at scale by neural models or the attacker targets on hacking specific questions of interest. To defend against such a threat, we discuss the necessity of building a misinformation-aware QA system that integrates question-answering and misinformation detection in a joint fashion.

Attacking Open-domain Question Answering by Injecting Misinformation

Previous attempts to incorporate a mention detection step into end-to-end neural coreference resolution for English have been hampered by a lack of singleton mention span data as well as other entity information. This paper presents a coreference model that learns singletons as well as features such as entity type and information status, via a multi-task learning-based approach. This approach achieves new state-of-the-art scores on the OntoGUM benchmark (+2.7 points) and increases robustness on multiple out-of-domain datasets (+2.3 points on average), likely due to greater generalizability for mention detection and utilization of more data from singletons, when compared to only coreferent mention pair matching.

Incorporating Singletons and Mention-based Features in Coreference Resolution via Multi-task Learning for Better Generalization

Over the past several years, demand and popularity of using virtual assistants to finish jobs like service scheduling and online shopping have increased. While keeping the user's request in mind, an effective task-oriented virtual agent must strive to improve the seller's profit. Therefore, in order to achieve the best possible trade-off between the parties, this form of virtual agents has to have strong negotiating abilities. Although current conversational agents are quite good at making fluent sentences, they are still unable to use strategic thinking. In order to more effectively contextualize the choice of the next set of negotiation methods while producing answers, we develop Nego-GAT, an end-to-end negotiation system that includes sentiment information and graph attention embedding into GPT-2. Our self-supervised model beats earlier cutting-edge negotiation models in terms of both the precision of strategy/dialogue act prediction and the caliber of the generated dialogue responses.

Sentiment Aided Graph Attentive Contextualization for Task Oriented Negotiation Dialogue Generation

Most benchmarks designed for question answering over knowledge bases (KBQA) operate with the i.i.d. assumption where one encounters the same schema items during inference as those observed during training. Recently, the GrailQA dataset was established to evaluate zero-shot generalization capabilities of KBQA models as a departure from the i.i.d. assumption. Reasonable performance of current KBQA systems on the zero-shot GrailQA split hints that the field might be moving towards more generalizable systems. In this work, we observe a bias in the GrailQA dataset towards simpler one or two-hop questions, which results in an inaccurate assessment of the aforementioned prowess. We propose GrailQA++, a challenging zero-shot KBQA test set that contains more questions relying on complex reasoning. We leverage the concept of graph isomorphisms to control the complexity of the questions and to ensure that our proposed test set has a fair distribution of simple and complex questions. Existing KBQA models suffer a substantial drop in performance on our constructed new test set as compared to the GrailQA zero-shot split. Our analysis reveals how isomorphisms can be used to understand the complementary strengths of different KBQA models and provide a deeper insight into model mispredictions. Overall, our paper highlights the \textit{non-generalizability} of existing models and the necessity for designing more challenging benchmarks. Our dataset is available at \url{https://github.com/sopankhosla/GrailQA-PlusPlus}

GrailQA++: A Challenging Zero-Shot Benchmark for Knowledge Base Question Answering

Large language models (LLMs) have demonstrated impressive capabilities in natural language generation. However, their output quality can be inconsistent, posing challenges for generating natural language from logical forms (LFs). This task requires the generated outputs to embody the exact semantics of LFs, without missing any LF semantics or creating any hallucinations. In this work, we tackle this issue by proposing a novel generate-and-rerank approach. Our approach involves initially generating a set of candidate outputs by prompting an LLM and subsequently reranking them using a task-specific reranker model. In addition, we curate a manually collected dataset to evaluate the alignment between different ranking metrics and human judgements. The chosen ranking metrics are utilized to enhance the training and evaluation of the reranker model. By conducting extensive experiments on three diverse datasets, we demonstrate that the candidates selected by our reranker outperform those selected by baseline methods in terms of semantic consistency and fluency, as measured by three comprehensive metrics. Our findings provide strong evidence for the effectiveness of our approach in improving the quality of generated outputs.

Premium content

SQUARE: Automatic Question Answering Evaluation using Multiple Positive and Negative References

Downloads

Next from IJCNLP-AACL 2023

MasakhaNEWS: News Topic Classification for African languages

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES