Canada

Conventional Transformer-based Video Ques-tion Answering (VideoQA) approaches gener-ally encode frames independently through oneor more image encoders followed by interac-tion between frames and question. However,such schema would incur significant memoryuse and inevitably slow down the training andinference speed. In this work, we present ahighly efficient approach for VideoQA basedon existing vision-language pre-trained mod-els where we concatenate video frames to an × n matrix and then convert it to one image.By doing so, we reduce the use of the imageencoder from n2 to 1 while maintaining thetemporal structure of the original video. Ex-perimental results on MSRVTT and TrafficQAshow that our proposed approach achievesstate-of-the-art performance with nearly 4×faster speed and only 30% memory use. Weshow that by integrating our approach intoVideoQA systems we can achieve compara-ble, even superior, performance with a sig-nificant speed up for training and inference.We believe the proposed approach can facil-itate VideoQA-related research by reducingthe computational requirements for those whohave limited access to budgets and resources. Our code will be made publicly available forresearch use.

ACL 2023

Is a Video worth n × n Images? A Highly Efficient Approach to Transformer-based Video Question Answering
Chenyang Lyu, Tianbo Ji, Yvette Graham and Jennifer Foster (chenyang.lyu2@mail.dcu.ie)
Status: Accept

vision question answering

vision transformer

multimodal

workshop paper

### Welcome to ACL 2023, the 61st Annual Meeting of the Association for Computational Linguistics! 
 The conference will be held in Toronto, Canada, July 9-14, 2023. 
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. 
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)**

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.

Scaling language models have revolutionized widespread NLP tasks, yet little comprehensively explored few-shot relation extraction with large language models. In this paper, we investigate in-context learning and data generation for few-shot relation extraction via GPT-3.5 through exhaustive experiments. To enhance few-shot performance, we further propose task-related instructions and schema-constrained data generation. We observe that in-context learning can achieve performance on par with previous prompt learning approaches, and data generation with the large language model can boost previous solutions to obtain new state-of-the-art few-shot results on four widely-studied relation extraction datasets.

How to Unleash the Power of Large Language Models for Few-shot Relation Extraction?
Xin Xu, Yuqi Zhu, Xiaohan Wang and Ningyu Zhang (xxucs@zju.edu.cn)
Status: Accept

Popular models for Knowledge Graph Question Answering (KGQA), including semantic parsing and End-to-End (E2E) models, decode into a constrained space of KG relations. Although E2E models accommodate novel entities at test-time, this constraint means they cannot access novel relations, requiring expensive and time-consuming retraining whenever a new relation is added to the KG. We propose KG-Flex, a new architecture for E2E KGQA that instead decodes into a continuous embedding space of relations, which enables use of novel relations at test-time. KG-Flex is the first to support KG updates with entirely novel triples, free of retraining, while still supporting end-to-end training with simple, weak supervision of (Q, A) pairs. Our architecture saves on time, energy, and data resources for retraining, yet we retain performance on standard benchmarks. We further demonstrate zero-shot use of novel relations, achieving up to 82\% of baseline hit@1 on three QA datasets. KG-Flex can also fine-tune, requiring significantly shorter time than full retraining; fine-tuning on target data for 10% of full training increases hit@1 to 89-100% of baseline.

KGQA Without Retraining
Nick McKenna and Priyanka Sen (nick.mckenna@ed.ac.uk)
Status: Accept

This paper investigates the problem of Named Entity Recognition (NER) for extreme low-resource languages with only a few hundred tagged data samples. NER is a fundamental task in Natural Language Processing (NLP). A critical driver accelerating NER systems' progress is the existence of large-scale language corpora that enable NER systems to achieve outstanding performance in languages such as English and French with abundant training data. However, NER for low-resource languages remains relatively unexplored. In this paper, we introduce Mask Augmented Named Entity Recognition (MANER), a new methodology that leverages the distributional hypothesis of pre-trained masked language models (MLMs) for NER. The <mask> token in pre-trained MLMs encodes valuable semantic contextual information. MANER re-purposes the <mask> token for NER prediction. Specifically, we prepend the <mask> token to every word in a sentence for which we would like to predict the named entity tag. During training, we jointly fine-tune the MLM and a new NER prediction head attached to each <mask> token. We demonstrate that MANER is well-suited for NER in low-resource languages; our experiments show that for 100 languages with as few as 100 training examples, it improves on state-of-the-art methods by up to 48% and by 12% on average on F1 score. We also perform detailed analyses and ablation studies to understand the scenarios that are best-suited to MANER.

MANER: Mask Augmented Named Entity Recognition for Extreme Low-Resource Languages
Shashank Sonkar, Zichao Wang and Richard Baraniuk (shashank.sonkar@rice.edu)
Status: Accept

Recently, compressive text summarisation offers a balance between the conciseness issue of extractive summarisation and the factual hallucination issue of abstractive summarisation. However, most existing compressive summarisation methods are supervised, relying on the expensive effort of creating a new training dataset with corresponding compressive summaries. In this paper, we propose an efficient and interpretable compressive summarisation method that utilises unsupervised dual-agent reinforcement learning to optimise a summary's semantic coverage and fluency by simulating human judgment on summarisation quality. Our model consists of an extractor agent and a compressor agent, and both agents have a multi-head attentional pointer-based structure. The extractor agent first chooses salient sentences from a document, and then the compressor agent compresses these extracted sentences by selecting salient words to form a summary without using reference summaries to compute the summary reward. To our best knowledge, this is the first work on unsupervised compressive summarisation. Experimental results on three widely used datasets (e.g., Newsroom, CNN/DM, and XSum) show that our model achieves promising performance and a significant improvement on Newsroom in terms of the ROUGE metric, as well as interpretability of semantic coverage of summarisation results.

Efficient and Interpretable Compressive Text Summarisation with Unsupervised Dual-Agent Reinforcement Learning
Peggy Tang, Junbin Gao, Lei Zhang and Zhiyong Wang (peggy.tang@sydney.edu.au)
Status: Accept

The recently introduced FNet model (Lee-Thorp et al., 2022) computes a two-dimensional discrete Fourier transform (DFT) of a sequence- length-by-hidden-dimension-sized representation of its input. Because it is equally efficient to compute the DFT of any reshaping of this input matrix, we investigate the extent to which increasing the frequency resolution in one dimension (at the expense of the other) affects task performance. We consider the LRA tasks (Tay et al., 2021) considered by Lee-Thorpe et al., as well as the more practical setting of using FNet as the encoder in a machine translation (MT) model. We find that frequency resolution has a marked task-dependent effect on performance, allowing us to largely outperform standard FNet on our tasks, and suggesting that resolution should be carefully tuned before deploying FNet.

Exploring the Effect of Frequency Resolution in FNet
Gregory Szumel, Ghazal Khalighinejad, Rickard Stureborg and Sam Wiseman (gks9@duke.edu)
Status: Accept

Interactive machine learning (IML) is a beneficial learning paradigm in cases of limited data availability, as human feedback is incrementally integrated into the training process. In this paper, we present an IML pipeline for image captioning which allows us to incrementally adapt a pre-trained image captioning model to a new data distribution based on user input. In order to incorporate user input into the model, we explore the use of a combination of simple data augmentation methods to obtain larger data batches for each newly annotated data instance and implement continual learning methods to prevent catastrophic forgetting from repeated updates. For our experiments, we split a domain-specific image captioning dataset, namely VizWiz, into non-overlapping parts to simulate an incremental input flow for continually adapting the model to new data. We find that, while data augmentation worsens results, even when relatively small amounts of data are available, episodic memory is an effective strategy to retain knowledge from previously seen clusters.

Towards Adaptable and Interactive Image Captioning with Data Augmentation and Episodic Memory
Aliki Anagnostopoulou, Mareike Hartmann and Daniel Sonntag (aliki.anagnostopoulou@dfki.de)
Status: Accept

Personalized dialogue agents (DAs) powered by large pre-trained language models (PLMs) often rely on explicit persona descriptions to maintain personality consistency. However, such descriptions may not always be available or may pose privacy concerns. To tackle this bottleneck, we introduce PersonaPKT, a lightweight transfer learning approach that can build persona-consistent dialogue models without explicit persona descriptions. By representing each persona as a continuous vector, PersonaPKT learns implicit persona-specific features directly from a small number of dialogue samples produced by the same persona, adding less than 0.1% trainable parameters for each persona on top of the PLM backbone. Empirical results demonstrate that PersonaPKT effectively builds personalized DAs with high storage efficiency, outperforming various baselines in terms of persona consistency while maintaining good response generation quality. In addition, it enhances privacy protection by avoiding explicit persona descriptions. Overall, PersonaPKT is an effective solution for creating personalized DAs that respect user privacy.

PersonaPKT: Building Personalized Dialogue Agents via Parameter-efficient Knowledge Transfer
Xu Han, Bin Guo, Yoon Jung, Benjamin Yao, Yu Zhang, Xiaohu Liu and Chenlei Guo (xuha2442@colorado.edu)
Status: Accept

Autocomplete is a task where the user inputs a piece of text, termed prompt, which is conditioned by the model to generate semantically coherent continuation. Existing works for this task have primarily focused on datasets (e.g., email, chat) with high frequency user prompt patterns (or focused prompts) where word-based language models have been quite effective. In this work, we study the more challenging open-domain setting consisting of low frequency user prompt patterns (or broad prompts, e.g., prompt about 93rd academy awards) and demonstrate the effectiveness of character-based language models. We study this problem under memory-constrained settings (e.g., edge devices and smartphones), where character-based representation is effective in reducing the overall model size (in terms of parameters). We use WikiText-103 benchmark to simulate broad prompts and demonstrate that character models rival word models in exact match accuracy for the autocomplete task, when controlled for the model size. For instance, we show that a 20M parameter character model performs similar to an 80M parameter word model in the vanilla setting. We further propose novel methods to improve character models by incorporating inductive bias in the form of compositional information and representation transfer from large word models.

Small Character Models Match Large Word Models for Autocomplete Under Memory Constraints
Ganesh Jawahar, Subhabrata Mukherjee, Debadeepta Dey, Muhammad Abdul-Mageed, Laks Lakshmanan, V.S., Caio Mendes, Gustavo de Rosa and Shital Shah (ganeshjwhr@gmail.com)
Status: Accept

The information retrieval community has made significant progress in improving the efficiency of Dual Encoder (DE) dense passage retrieval systems, making them suitable for latency-sensitive settings. However, many proposed procedures are often too complex or resource-intensive, which makes it difficult for practitioners to adopt them or identify sources of empirical gains. Therefore, in this work, we propose a trivially simple recipe to serve as a baseline method for boosting the efficiency of DE retrievers leveraging an asymmetric architecture. Our results demonstrate that even a 2-layer, BERT-based query encoder can still retain 92.5% of the full DE performance on the BEIR benchmark via unsupervised distillation and proper student initialization. We hope that our findings will encourage the community to re-evaluate the trade-offs between method complexity and performance improvements.

Query Embedding Alignment: a Trivially Simple Recipe to Boost Dense Passage Retrieval Efficiency
Yuxuan Wang and Lyu Hong (wangy49@seas.upenn.edu)
Status: Accept

For many languages and applications, even though enough data is available for training Named Entity Disambiguation (NED) systems, few off-the-shelf models are available for use in practice. This is due to both the large size of state-of-the-art models, and to the computational requirements for recreating them from scratch. However, we observe that in practice, acceptable models can be trained and run with far fewer resources. In this work, we intro- duce MiniNED, a framework for creating small NED models from medium-sized datasets. The resulting models can be tuned for application- specific objectives and trade-offs, depending on practitioners’ requirements concerning model size, frequency bias, and out-of-domain gener- alization. We evaluate the framework in nine languages, and achieve reasonable performance using models that are a fraction of the size of recent work.

Downloads

Next from ACL 2023

How to Unleash the Power of Large Language Models for Few-shot Relation Extraction? Xin Xu, Yuqi Zhu, Xiaohan Wang and Ningyu Zhang (xxucs@zju.edu.cn) Status: Accept

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

.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