United Arab Emirates

Contextualized word embeddings have emerged as the most important tool for performing NLP tasks in a large variety of languages. In order to improve the cross- lingual representation and transfer learning quality, contextualized embedding alignment techniques, such as mapping and model fine-tuning, are employed. Existing techniques however are time-, data- and computational resource-intensive. In this paper we analyze these techniques by utilizing three tasks: bilingual lexicon induction (BLI), word retrieval and cross-lingual natural language inference (XNLI) for a high resource (German-English) and a low resource (Bengali-English) language pair. In contrast to previous works which focus only on a few popular models, we compare five multilingual and seven monolingual language models and investigate the effect of various aspects on their performance, such as vocabulary size, number of languages used for training and number of parameters. Additionally, we propose a parameter-, data- and runtime-efficient technique which can be trained with 10% of the data, less than 10% of the time and have less than 5% of the trainable parameters compared to model fine-tuning. We show that our proposed method is competitive with resource heavy models, even outperforming them in some cases, even though it relies on less resource

EMNLP 2022

Comparative Analysis of Cross-lingual Contextualized Word Embeddings

contextualized embeddings alignment

multilinguality

workshop paper

## 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.

As emotion analysis in text has gained a lot of attention in the field of natural language processing, differences in emotion expression across languages could have consequences for how emotion detection models work. We evaluate the language-dependence of an mBERT-based emotion detection model by comparing language identification performance before and after fine-tuning on emotion detection, and performing (adjusted) zero-shot experiments to assess whether emotion detection models rely on language-specific information. When dealing with typologically dissimilar languages, we found evidence for the language-dependence of emotion detection.

How Language-Dependent is Emotion Detection? Evidence from Multilingual BERT

BERT-style contextualized word embedding models are critical for good performance in most NLP tasks, but they are data-hungry and therefore difficult to train for low-resource languages. In this work, we investigate whether a combination of greatly reduced model size and two linguistically rich auxiliary pretraining tasks (part-of-speech tagging and dependency parsing) can help produce better BERTs in a low-resource setting. Results from 7 diverse languages indicate that our model, MicroBERT, is able to produce marked improvements in downstream task evaluations, including gains up to 18% for parser LAS and 11% for NER F1 compared to an mBERT baseline, and we achieve these results with less than 1% of the parameter count of a multilingual BERT base--sized model. We conclude that training very small BERTs and leveraging any available labeled data for multitask learning during pretraining can produce models which outperform both their multilingual counterparts and traditional fixed embeddings for low-resource languages.

MicroBERT: Effective Training of Low-resource Monolingual BERTs through Parameter Reduction and Multitask Learning

We present the University of British Columbia's submission to the MRL shared task on multilingual clause-level morphology. Our submission extends word-level inflectional models to the clause-level in two ways: first, by evaluating the role that BPE has on the learning of inflectional morphology, and second, by evaluating the importance of a copy bias obtained through data hallucination. Experiments demonstrate a strong preference for language-tuned BPE and a copy bias over a vanilla transformer. The methods are complementary for inflection and analysis tasks – combined models see error reductions of 38% for inflection and 15.6% for analysis; However, this synergy does not hold for reinflection, which performs best under a BPE-only setting. A deeper analysis of the errors generated by our models illustrates that the copy bias may be too strong - the combined model produces predictions more similar to the copy-influenced system, despite the success of the BPE-model.

Impact of Sequence Length and Copying on Clause-Level Inflection

This position paper discusses the problem of multilingual evaluation. Using simple statistics, such as average language performance, might inject linguistic biases in favor of dominant language families into evaluation methodology. We argue that a qualitative analysis informed by comparative linguistics is needed for multilingual results to detect this kind of bias. We show in our case study that results in published works can indeed be linguistically biased and we demonstrate that visualization based on URIEL typological database can detect it.

Average Is Not Enough: Caveats of Multilingual Evaluation

This paper presents a new natural language processing task - Actionable Entities Recognition (AER) - recognition of entities that protagonists could interact with for further plot development. Though similar to classical Named Entity Recognition (NER), it has profound differences. In particular, it is crucial for interactive fiction, where the agent needs to detect entities that might be useful in the future. We also discuss if AER  might be further useful for the systems dealing with narrative processing since actionable entities profoundly impact the causal relationship in a story. We validate the proposed task on two previously available datasets and present a new benchmark dataset for the AER task that includes 5550 descriptions with one or more actionable entities. 

AERBIF: Actionable Entities Recognition Benchmark for Interactive Fiction

Task planning is an important component of traditional robotics systems enabling robots to compose fine grained skills to perform more complex tasks. Recent work building systems for translating natural language to executable actions for task completion in simulated embodied agents is focused on directly predicting low level action sequences that would be expected to be directly executable by a physical robot. In this work, we instead focus on predicting a higher level plan representation for one such embodied task completion dataset - TEACh, under the assumption that techniques for high-level plan prediction from natural language are expected to be more transferable to physical robot systems . We demonstrate that better plans can be predicted using multimodal context, and that plan prediction and plan execution modules are likely dependent on each other and hence it may not be ideal to fully decouple them. Further, we benchmark execution of oracle plans to quantify the scope for improvement in plan prediction models.

Multimodal Contextualized Plan Prediction for Embodied Task Completion

We design and evaluate a Thompson sampling-based interactive Bayesian optimization algorithm that leverages a Gaussian process reward model of the Masked Language Model (MLM) pre-training objective, for its sequential minimization. Transformer-based language model (TLM) pre-training requires large volumes of data and high computational resources, while introducing many unresolved design choices, such as hyperparameter selection of the pre-training procedure. We here fit TLM pre-training validation losses with a Gaussian process, and formulate a Thompson sampling bandit policy that maximizes its sequentially attained cumulative rewards. Instead of MLM pre-training with fixed masking probabilities, the proposed Gaussian process-based Thompson sampling (GP-TS) accelerates and improves MLM pre-training performance by sequentially selecting masking hyperparameters of the language model. GP-TS provides an interactive, efficient framework for pre-training TLMs, as it attains better MLM pre-training loss in less epochs, avoiding costly hyperparameter selection techniques.

Thompson sampling for interactive Bayesian optimization of dynamic masking-based language model pre-training

Recently, there has been an increasing inter- est in two-pass streaming end-to-end speech recognition (ASR) that incorporates a 2nd-pass rescoring model on top of the conventional 1st- pass streaming ASR model to improve recogni- tion accuracy while keeping latency low. One of the latest 2nd-pass rescoring model, Trans- former Rescorer, takes the n-best initial out- puts and audio embeddings from the 1st-pass model, and then choose the best output by re- scoring the n-best initial outputs. However, training this Transformer Rescorer requires ex- pensive paired audio-text training data because the model uses audio embeddings as input. In this work, we present our Joint Audio/Text training method for Transformer Rescorer, to leverage unpaired text-only data which is rela- tively cheaper than paired audio-text data. We evaluate Transformer Rescorer with our Joint Audio/Text training on Librispeech dataset as well as our large-scale in-house dataset and show that our training method can improve word error rate (WER) significantly compared to standard Transformer Rescorer without re- quiring any extra model parameters or latency.

Joint Audio/Text Training for Transformer Rescorer of Streaming Speech Recognition

Text games present opportunities for natural language understanding (NLU) methods to tackle reinforcement learning (RL) challenges. However, recent work has questioned the necessity of NLU by showing random text hashes could perform decently. In this paper, we pursue a fine-grained investigation into the roles of text in the face of different RL challenges, and reconcile that semantic and non-semantic language representations could be complementary rather than contrasting. Concretely, we propose a simple scheme to extract relevant contextual information into an approximate state hash as extra input for an RNN-based text agent. Such a lightweight plug-in achieves competitive performance with state-of-the-art text agents using advanced NLU techniques such as knowledge graph and passage retrieval, suggesting non-NLU methods might suffice to tackle the challenge of partial observability. However, if we remove RNN encoders and use approximate or even ground-truth state hash alone, the model performs miserably, which confirms the importance of semantic function approximation to tackle the challenge of combinatorially large observation and action spaces. Our findings and analysis provide new insights for designing better text game task setups and agents.

Revisiting the Roles of "Text” in Text Games

Understanding rich narratives, such as dialogues and stories, often requires natural language processing systems to access relevant knowledge from commonsense knowledge graphs. However, these systems typically retrieve facts from KGs using simple heuristics that disregard the complex challenges of identifying situationally-relevant commonsense knowledge (e.g., contextualization, implicitness, ambiguity). In this work, we propose the new task of commonsense fact linking, where models are given contexts and trained to identify situationally-relevant commonsense knowledge from KGs. Our novel benchmark, ComFact, contains ~293k in-context relevance annotations for commonsense triplets across four stylistically diverse dialogue and storytelling datasets. Experimental results confirm that heuristic fact linking approaches are imprecise knowledge extractors. Learned fact linking models demonstrate across-the-board performance improvements (~34.6% F1) over these heuristics. Furthermore, improved knowledge retrieval yielded average downstream improvements of 9.8% for a dialogue response generation task. However, fact linking models still significantly underperform humans, suggesting our benchmark is a promising testbed for research in commonsense augmentation of NLP systems.

Downloads

Next from EMNLP 2022

How Language-Dependent is Emotion Detection? Evidence from Multilingual BERT

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

Next from EMNLP 2022

How Language-Dependent is Emotion Detection? Evidence from Multilingual BERT

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

Downloads