Canada

In question answering requiring common sense, language models (e.g., GPT-3) have been used to generate text expressing background knowledge that helps improve performance. Yet the cost of working with such models is very high; in this work, we finetune smaller language models to generate useful intermediate context, referred to here as elaborations. Our framework alternates between updating two language models---an elaboration generator and an answer predictor---allowing each to influence the other. Using less than 0.5\% of the parameters of GPT-3, our model outperforms alternatives with similar sizes and closes the gap with GPT-3 on four commonsense question answering benchmarks. Human evaluations show that the quality of the generated elaborations is high.

&lt;iframe src=&quot;https://app.sli.do/event/qQ87ZDnDYN6cfmCQWHFaok/embed/polls/5d2d11eb-3a16-4308-8dda-d5ba354a8537&quot; width=&quot;300&quot; height=&quot;400&quot;&gt;&lt;/iframe&gt;

ACL 2023

Elaboration-Generating Commonsense Question Answering at Scale

language generation

commonsense

question answering

In question answering requiring common sense, language models (e.g., GPT-3) have been used to generate text expressing background knowledge that helps improve performance. Yet the cost of working with such models is very high; in this work, we finetune smaller language models to generate useful intermediate context, referred to here as elaborations. Our framework alternates between updating two language models---an elaboration generator and an answer predictor---allowing each to influence the other. Using less than 0.5\% of the parameters of GPT-3, our model outperforms alternatives with similar sizes and closes the gap with GPT-3 on four commonsense question answering benchmarks. Human evaluations show that the quality of the generated elaborations is high.

<iframe src="https://app.sli.do/event/qQ87ZDnDYN6cfmCQWHFaok/embed/polls/5d2d11eb-3a16-4308-8dda-d5ba354a8537" width="300" height="400"></iframe>

poster

### Welcome to ACL 2023, the 61st Annual Meeting of the Association for Computational Linguistics! 
<br>The conference will be held in Toronto, Canada, July 9-14, 2023.<br><br>
Following the succession of the recent conferences in our field, ACL 2023 will adopt a hybrid format.
While the impact of Covid has considerably diminished in terms of traveling, obtaining visas to Canada
entails a very long process. Moreover, the global economic conditions pose challenges for many individuals to travel to conferences. Recognizing these circumstances, we know many participants may not be
able to attend the conference in person. Therefore, we are committed to providing a great virtual platform
so everyone has the opportunity to interact with other participants and enjoy the conference. Based on the
current registered participants, approxiately 30% have chosen to attend the conference virtually. Whether
you join us in person or virtually, we sincerely hope everyone has a remarkable conference experience.<br>
This General Chair’s message is where I express my gratitude to the many individuals who have made
enormous contributions to the conference over the past year.

Read [**ACL 2023 General Chair's message**](https://docs.google.com/document/d/1WobYM7norbG4dI48s75HfJoD89qgX5a_F-6U8AteLSA/edit?usp=sharing/) in full.

##### **[Conference Handbook](https://2023.aclweb.org/downloads/acl2023-handbook.pdf)**
<br><br><br><br>

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.

Syntactic probing methods have been used to examine whether and how pre-trained language models (PLMs) encode syntactic features. However, the probing methods are usually biased by the PLMs' memorization of common word co-occurrences, even if they do not form syntactic relations. This paper presents a random-word-substitution and random-label-matching control task to reduce these biases and improve the robustness of syntactic probing methods. Our control tasks are also shown to notably improve the consistency of probing results between different probing methods and make the methods more robust with respect to the text attributes of the probing instances. Our control tasks make syntactic probing methods better at reconstructing syntactic features and more generalizable to unseen text domains. Our experiments show that our proposed control tasks are effective on different PLMs, probing methods, and syntactic features.

<iframe src="https://app.sli.do/event/qQ87ZDnDYN6cfmCQWHFaok/embed/polls/2f66252b-d2de-474a-8384-0e96413c7e3b" width="300" height="400"></iframe>

Improving Syntactic Probing Correctness and Robustness with Control Tasks

Large-scale language model pretraining is a very successful form of self-supervised learning in natural language processing, but it is increasingly expensive to perform as the models and pretraining corpora have become larger over time. We propose NarrowBERT, a modified transformer encoder that increases the throughput for masked language model pretraining by more than 2x. NarrowBERT sparsifies the transformer model such that the self-attention queries and feedforward layers only operate on the masked tokens of each sentence during pretraining, rather than all of the tokens as with the usual transformer encoder. We also show that NarrowBERT increases the throughput at inference time by as much as 3.5x with minimal (or no) performance degradation on sentence encoding tasks like MNLI. Finally, we examine the performance of NarrowBERT on the IMDB and Amazon reviews classification and CoNLL NER tasks and show that it is also comparable to standard BERT performance.

<iframe src="https://app.sli.do/event/qQ87ZDnDYN6cfmCQWHFaok/embed/polls/e0ea274e-7ade-476a-afed-03c012cb1388" width="300" height="400"></iframe>

NarrowBERT: Accelerating Masked Language Model Pretraining and Inference

Pre-trained language models have been successful in many knowledge-intensive NLP tasks. However, recent work has shown that models such as BERT are not ``structurally ready'' to aggregate textual information into a [CLS] vector for dense passage retrieval (DPR). This ``lack of readiness'' results from the gap between language model pre-training and DPR fine-tuning.  Previous solutions call for computationally expensive techniques such as hard negative mining, cross-encoder distillation, and further pre-training to learn a robust DPR model. In this work, we instead propose to fully exploit knowledge in a pre-trained language model for DPR by aggregating the contextualized token embeddings into a dense vector, which we call agg*.  By concatenating vectors from the [CLS] token and agg*, our Aggretriever model substantially improves the effectiveness of dense retrieval models on both in-domain and zero-shot evaluations without introducing substantial training overhead. Code is available at https://github.com/castorini/dhr

Aggretriever: A Simple Approach to Aggregate Textual Representations for Robust Dense Passage Retrieval

Recent work in natural language processing (NLP) has yielded appealing results from scaling; however, using only scale to improve performance means that resource consumption also scales. Resources include data, time, storage, or energy, all of which are naturally limited and unevenly distributed. This motivates research into efficient methods that require fewer resources to achieve similar results. This survey synthesizes and relates current methods and findings in efficient NLP. We aim to provide both guidance for conducting NLP under limited resources, and point towards promising research directions for developing more efficient methods.

Efficient Methods for Natural Language Processing: A Survey

This work introduces the D-WISE Tool Suite (DWTS), a novel working environment for digital qualitative discourse analysis in the Digital Humanities (DH). The DWTS addresses limitations of current DH tools induced by the ever-increasing amount of heterogeneous, unstructured, and multi-modal data in which the discourses of contemporary societies are encoded. To provide meaningful insights from such data, our system leverages and combines state-of-the-art machine learning technologies from Natural Language Processing and Com-
puter Vision. Further, the DWTS is conceived and developed by an interdisciplinary team of
cultural anthropologists and computer scientists to ensure the tool's usability for modern
DH research. Central features of the DWTS are: a) import of multi-modal data like text, image, audio, and video b) preprocessing pipelines for automatic annotations c) lexical and semantic search of documents d) manual span, bounding box, time-span, and frame annotations e) documentation of the research process.

<iframe src="https://app.sli.do/event/qQ87ZDnDYN6cfmCQWHFaok/embed/polls/0c460713-8008-4a92-85d5-b7758c2d4669" width="300" height="400"></iframe>

The D-WISE Tool Suite: Multi-Modal Machine-Learning-Powered Tools Supporting and Enhancing Digital Discourse Analysis

The Zero-Shot Learning (ZSL) task pertains to the identification of entities or relations in texts that were not seen during training. ZSL has emerged as a critical research area due to the scarcity of labeled data in specific domains, and its applications have grown significantly in recent years. With the advent of large pretrained language models, several novel methods have been proposed, resulting in substantial improvements in ZSL performance. There is a growing demand, both in the research community and industry, for a comprehensive ZSL framework that facilitates the development and accessibility of the latest methods and pretrained models.
In this study, we propose a novel ZSL framework called Zshot that aims to address the aforementioned challenges. Our primary objective is to provide a platform that allows researchers to compare different state-of-the-art ZSL methods with standard benchmark datasets. Additionally, we have designed our framework to support the industry with readily available APIs for production under the standard SpaCy NLP pipeline. Our API is extendible and evaluable, moreover, we include numerous enhancements such as boosting the accuracy with pipeline ensembling and visualization utilities available as a SpaCy extension.

<iframe src="https://app.sli.do/event/uyG7EHDP5trs2LHubv2Gj8/embed/polls/3a06179a-2fe5-465a-a1c3-4a848acba583" width="300" height="400"></iframe>

Zshot: An Open-source Framework for Zero-Shot Named Entity Recognition and Relation Extraction

Heterogenity of data-to-text generation datasets limits the research on data-to-text generation systems. We present TabGenie -- a toolkit which enables researchers to explore, preprocess, and analyze a variety of data-to-text generation datasets through the unified framework of table-to-text generation. In TabGenie, all inputs are represented as tables with associated metadata. The tables can be explored through a web interface, which also provides an interactive mode for debugging table-to-text generation, facilitates side-by-side comparison of generated system outputs, and allows easy exports for manual analysis. Furthermore, TabGenie is equipped with command line processing tools and Python bindings for unified dataset loading and processing. We release TabGenie as a PyPI package and provide its open-source code and a live demo at https://github.com/kasnerz/tabgenie.

<iframe src="https://app.sli.do/event/qQ87ZDnDYN6cfmCQWHFaok/embed/polls/e34d0717-37b4-44ba-9d5c-71030a74b592" width="300" height="400"></iframe>

TabGenie: A Toolkit for Table-to-Text Generation

Online conversation is a ubiquitous way to share information and connect everyone but repetitive idiomatic text typing takes users a lot of time.
This paper demonstrates a simple yet effective cloud based smart compose system to improve human-to-human conversation efficiency. 
Heuristics from different perspectives are designed to achieve the best trade-off between quality and latency.
From the modeling side, the decoder-only model exploited the previous turns of conversational history in a computation lightweight manner. 
Besides, a novel phrase tokenizer is proposed to reduce latency without losing the composing quality further. 
Additionally, the caching mechanism is applied to the serving framework. 
The demo video of the system is available at https://youtu.be/U1KXkaqr60g.
We open-sourced our phrase tokenizer in https://github.com/tensorflow/text.

<iframe src="https://app.sli.do/event/qQ87ZDnDYN6cfmCQWHFaok/embed/polls/2055766b-179d-4a6d-bb6c-9d257fd6a489" width="300" height="400"></iframe>

An Efficient Conversational Smart Compose System

We present KWJA, a high-performance unified Japanese text analyzer based on foundation models.
KWJA supports a wide range of tasks, including typo correction, word segmentation, word normalization, morphological analysis, named entity recognition, linguistic feature tagging, dependency parsing, PAS analysis, bridging reference resolution, coreference resolution, and discourse relation analysis, making it the most versatile among existing Japanese text analyzers.
KWJA solves these tasks in a multi-task manner but still achieves competitive or better performance compared to existing analyzers specialized for each task.
KWJA is publicly available under the MIT license at https://github.com/ku-nlp/kwja.

<iframe src="https://app.sli.do/event/qQ87ZDnDYN6cfmCQWHFaok/embed/polls/c676b4bb-ccd1-4e02-baf8-9bdeff179c50" width="300" height="400"></iframe>

KWJA: A Unified Japanese Analyzer Based on Foundation Models

An open-source DeepPavlov Dream Platform is specifically tailored for development of complex dialog systems like Generative AI Assistants. The stack prioritizes efficiency, modularity, scalability, and extensibility with the goal to make it easier to develop complex dialog systems from scratch. It supports modular approach to implementation of conversational agents enabling their development through the choice of NLP components and conversational skills from a rich library organized into the distributions of ready-for-use multi-skill AI assistant systems. In DeepPavlov Dream, multi-skill Generative AI Assistant consists of NLP components that extract features from user utterances, conversational skills that generate or retrieve a response, skill and response selectors that facilitate choice of relevant skills and the best response, as well as a conversational orchestrator that enables creation of multi-skill Generative AI Assistants scalable up to industrial grade AI assistants. The platform allows to integrate large language models into dialog pipeline, customize with prompt engineering, handle multiple prompts during the same dialog session and create simple multimodal assistants.

<iframe src="https://app.sli.do/event/qQ87ZDnDYN6cfmCQWHFaok/embed/polls/c6c8c00c-0795-4677-b219-271531a70968" width="300" height="400"></iframe>

Premium content

Elaboration-Generating Commonsense Question Answering at Scale

Downloads

Next from ACL 2023

Improving Syntactic Probing Correctness and Robustness with Control Tasks

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES