Malta

Accurately understanding temporal relations between events is a critical building block of diverse tasks, such as temporal reading comprehension (TRC) and relation extraction (TRE). For example in TRC, we need to understand the temporal semantic differences between the following two questions that are lexically near-identical: “What finished right before the decision?” or “What finished right after the decision?”. To discern the two questions, existing solutions have relied on answer overlaps as a proxy label to contrast similar and dissimilar questions. However, we claim that answer overlap can lead to unreliable results, due to spurious overlaps of two dissimilar questions with coincidentally identical answers. To address the issue, we propose a novel approach that elicits proper reasoning behaviors through a module for predicting time spans of events. We introduce the Timeline Reasoning Network (TRN) operating in a two-step inductive reasoning process: In the first step model initially answers each question with semantic and syntactic information. The next step chains multiple questions on the same event to predict a timeline, which is then used to ground the answers. Results on the TORQUE and TB-dense, TRC, and TRE tasks respectively, demonstrate that TRN outperforms previous methods by effectively resolving the spurious overlaps using the predicted timeline.

EACL 2024

Chaining Event Spans for Temporal Relation Grounding

inductive reasoning

timeline

spurious correlation

temporal reasoning

technical paper

Welcome to the 18th Conference of the European Chapter of the Association for Computational Linguistics (EACL). Continuing its mission of expanding and involving the science community of all European countries, EACL has selected the Malta community for the 18th EACL. Considering the importance of physical interaction among researchers, the conference will be held at the Hotel Radisson Blu, St. Julians, in Malta, from 17 to 22 of March, 2024. The conference will also feature the possibility to participate virtually. As the flagship European conference in the field of computational linguistics, EACL welcomes European and international researchers covering a broad spectrum of research areas that are concerned with computational approaches to natural language. 
#### [Conference Handbook](https://drive.google.com/file/d/13Wn38q0yev6U3RTgnf4Tn5T9GuAuZ-xW/view?usp=sharing) *downloadable*

To access the EACL 2024 event page you need to register [**here**](https://acl.swoogo.com/EACL2024). 

Please register!

As the flagship European conference in the field of computational linguistics, EACL welcomes European and international researchers covering a broad spectrum of research areas that are concerned with computational approaches to natural language.

Recent studies of the emergent capabilities of transformer-based Natural Language Understanding (NLU) models have indicated that they have an understanding of lexical and compositional semantics. We provide evidence that suggests these claims should be taken with a grain of salt: we find that state-of-the-art Natural Language Inference (NLI) models are sensitive towards minor semantics preserving surface-form variations, which lead to sizable inconsistent model decisions during inference. Notably, this behaviour differs from valid and in-depth comprehension of compositional semantics, however does neither emerge when evaluating model accuracy on standard benchmarks nor when probing for syntactic, monotonic, and logically robust reasoning. We propose a novel framework to measure the extent of semantic sensitivity. To this end, we evaluate NLI models on adversarially generated examples containing minor semantics-preserving surface-form input noise. This is achieved using conditional text generation, with the explicit condition that the NLI model predicts the relationship between the original and adversarial inputs as a symmetric equivalence entailment. We systematically study the effects of the phenomenon across NLI models for $\text{\emph{in-}}$ and $\text{\emph{out-of-}}$ domain settings. Our experiments show that semantic sensitivity causes performance degradations of $12.92\%$ and $23.71\%$ average over $\text{\emph{in-}}$ and $\text{\emph{out-of-}}$ domain settings, respectively. We further perform ablation studies, analysing this phenomenon across models, datasets, and variations in inference and show that semantic sensitivity can lead to major inconsistency within model predictions.

Semantic Sensitivities and Inconsistent Predictions: Measuring the Fragility of NLI Models

We explore how weak supervision on abundant unlabeled data can be leveraged to improve few-shot performance in aspect-based sentiment analysis (ABSA) tasks. We propose a pipeline approach to construct a noisy ABSA dataset, and we use it to adapt a pre-trained sequence-to-sequence model to the ABSA tasks. We test the resulting model on three widely used ABSA datasets, before and after fine-tuning. Our proposed method preserves the full fine-tuning performance while showing significant improvements (15.84 absolute F1) in the few-shot learning scenario for the harder tasks. In zero-shot (i.e., without fine-tuning), our method outperforms the previous state of the art on the aspect extraction sentiment classification (AESC) task and is, additionally, capable of performing the harder aspect sentiment triplet extraction (ASTE) task.

A Weak Supervision Approach for Few-Shot Aspect Based Sentiment Analysis

We present Text-to-OverpassQL, a task designed to facilitate a natural language interface for querying geodata from OpenStreetMap (OSM). The Overpass Query Language (OverpassQL) allows users to formulate complex database queries and is widely adopted in the OSM ecosystem. Generating Overpass queries from natural language input serves multiple use-cases. It enables novice users to utilize OverpassQL without prior knowledge, assists experienced users with crafting advanced queries, and enables tool-augmented large language models to access information stored in the OSM database. In order to assess the performance of current sequence generation models on this task, we propose OverpassNL, a dataset of 8,352 queries with corresponding natural language inputs. We further introduce task specific evaluation metrics and ground the evaluation of the Text-to-OverpassQL task by executing the queries against the OSM database. We establish strong baselines by finetuning sequence-to-sequence models and adapting large language models with in-context examples. The detailed evaluation reveals strengths and weaknesses of the considered learning strategies, laying the foundations for further research into the Text-to-OverpassQL task.

Text-to-OverpassQL: A Natural Language Interface for Complex Geodata Querying of OpenStreetMap

In this paper, we introduce UNSEE, which stands for Unsupervised Non-Contrastive Sentence Embeddings. UNSEE demonstrates better performance compared to SimCSE in the Massive Text Embedding (MTEB) benchmark. We begin by highlighting the issue of representation collapse that occurs with the replacement of contrastive objectives with non-contrastive objectives in SimCSE. Subsequently, we introduce a straightforward solution called the target network to mitigate this problem. This approach enables us to harness non-contrastive objectives while ensuring training stability and achieving performance improvements similar to those seen with contrastive objectives. We have reached peak performance in non-contrastive sentence embeddings through extensive fine-tuning and optimization. These efforts have resulted in superior sentence representation models, emphasizing the importance of careful tuning and optimization for non-contrastive objectives.

UNSEE: Unsupervised Non-contrastive Sentence Embeddings

Translated texts bear several hallmarks distinct from texts originating in the language (“translationese”). Though individual translated texts are often fluent and preserve meaning, at a large scale, translated texts have statistical tendencies which distinguish them from text originally written in the language and can affect model performance. We frame the novel task of translationese reduction and hypothesize that Abstract Meaning Representation (AMR), a graph-based semantic representation which abstracts away from the surface form, can be used as an interlingua to reduce the amount of translationese in translated texts. By parsing English translations into an AMR and then generating text from that AMR, the result more closely resembles originally English text across three quantitative macro-level measures, without severely compromising fluency or adequacy. We compare our AMR-based approach against three other techniques based on machine translation or paraphrase generation. This work represents the first approach to reducing translationese in text and highlights the promise of AMR, given that our AMR-based approach outperforms more computationally intensive methods.

Lost in Translationese? Reducing Translation Effect Using Abstract Meaning Representation

How can NLP/AI practitioners engage with oral societies and develop locally appropriate language technologies? We report on our experience of working together over five years in a remote community in the far north of Australia, and how we prototyped simple language technologies to support our collaboration. We navigated different understandings of language, the functional differentiation of oral vs institutional languages, and the distinct technology opportunities for each. Our collaboration unsettled the first author's western framing of language as data for exploitation by machines, and we devised a design pattern that seems better aligned with local interests and aspirations. We call for new collaborations on the design of locally appropriate technologies for languages with primary orality.

Centering the Speech Community

Indigenous languages are historically under-served by Natural Language Processing (NLP) technologies, but this is changing for some languages with the recent scaling of large multilingual models and an increased focus by the NLP community on endangered languages. This position paper explores ethical considerations in building NLP technologies for Indigenous languages, based on the premise that such projects should primarily serve Indigenous communities. We report on interviews with 17 researchers working in or with Aboriginal and/or Torres Strait Islander communities on language technology projects in Australia. Drawing on insights from the interviews, we recommend practices for NLP researchers to increase attention to the process of engagements with Indigenous communities, rather than focusing only on decontextualised artefacts.

``It's how you do things that matters'': Attending to Process to Better Serve Indigenous Communities with Language Technologies

This paper explores the moral judgment and moral reasoning abilities exhibited by Large Language Models (LLMs) across languages through the Defining Issues Test. It is a well known fact that moral judgment depends on the language in which the question is asked. We extend the work of beyond English, to 5 new languages (Chinese, Hindi, Russian, Spanish and Swahili), and probe three LLMs -- ChatGPT, GPT-4 and Llama2Chat-70B -- that shows substantial multilingual text processing and generation abilities. Our study shows that the moral reasoning ability for all models, as indicated by the post-conventional score, is substantially inferior for Hindi and Swahili, compared to Spanish, Russian, Chinese and English, while there is no clear trend for the performance of the latter four languages. The moral judgments too vary considerably by the language.

Do Moral Judgment and Reasoning Capability of LLMs Change with Language? A Study using the Multilingual Defining Issues Test

The labor market is changing rapidly, prompting increased interest in the automatic extraction of occupational skills from text. With the advent of English benchmark job description datasets, there is a need for systems that handle their diversity well. We tackle the complexity in occupational skill datasets tasks---combining and leveraging multiple datasets for skill extraction, to identify rarely observed skills within a dataset, and overcoming the scarcity of skills across datasets. In particular, we investigate the retrieval-augmentation of language models, employing an external datastore for retrieving similar skills in a dataset-unifying manner. Our proposed method, \textbf{N}earest \textbf{N}eighbor \textbf{O}ccupational \textbf{S}kill \textbf{E}xtraction (NNOSE) effectively leverages multiple datasets by retrieving neighboring skills from other datasets in the datastore. This improves skill extraction \emph{without} additional fine-tuning. Crucially, we observe a performance gain in predicting infrequent patterns, with substantial gains of up to 30\% span-F1 in cross-dataset settings.

NNOSE: Nearest Neighbor Occupational Skill Extraction

Large Foundational Language Models are capable of performing many tasks at a high level but are difficult to deploy in many applications because of their size and proprietary ownership. Many will be motivated to distill specific capabilities of foundational models into smaller models that can be owned and controlled. In the development of a therapeutic chatbot, we wish to distill a capability known as reflective listening, in which a therapist produces reflections of client speech. These reflections either restate what a client has said, or connect what was said to a relevant observation, idea or guess that encourages and guides the client to continue contemplation. In this paper, we present a method for distilling the generation of reflections from a Foundational Language Model (GPT-4) into smaller models. We first show that GPT-4, using zero-shot prompting, can generate reflections at near 100% success rate, superior to all previous methods. Using reflections generated by GPT-4, we fine-tune different sizes of the GPT-2 family. The GPT-2-small model achieves 83% success on a hold-out test set and the GPT-2 XL achieves 90% success. We also show that GPT-4 can help in the labor-intensive task of evaluating the quality of the distilled models, using it as a zero-shot classifier. Using triple-human review as a guide, the classifier achieves a Cohen-Kappa of 0.66, a substantial inter-rater reliability figure.

Premium content

Chaining Event Spans for Temporal Relation Grounding

Downloads

Next from EACL 2024

Semantic Sensitivities and Inconsistent Predictions: Measuring the Fragility of NLI Models

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES