Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop)

38 papers
Feriji: A French-Zarma Parallel Corpus, Glossary & Translator (2024.acl-srw)

Copied to clipboard

Challenge: MT has seen significant advances in recent years, but the representation of African languages in MT systems is underrepresented due to linguistic complexities and limited resources.
Approach: They propose a first robust parallel French-Zarma corpus and a glossary for MT that contains 61,085 sentences in Zarma and 42,789 in French.
Outcome: The proposed model improves the representation of the Zarma language, a dialect of Songhay, spoken by over 5 million people across Niger and neighboring countries.
Pragmatic inference of scalar implicature by LLMs (2024.acl-srw)

Copied to clipboard

Challenge: Existing Large Language Models (LLMs) engage in pragmatic inference of scalar implicature, such as some.
Approach: They investigate how Large Language Models (LLMs) engage in pragmatic inference of scalar implicature, such as some.
Outcome: The proposed models interpret some as pragmatic implicature not all in the absence of context, aligning with human language processing.
Topic Modeling for Short Texts with Large Language Models (2024.acl-srw)

Copied to clipboard

Challenge: Large Language Models (LLMs) can be used to solve topic modeling challenges for short texts by contextually learning the meanings of words.
Approach: They propose two approaches to using Large Language Models (LLMs) for topic modeling: parallel prompting and sequential prompting.
Outcome: The proposed methods identify more coherent topics than existing ones while maintaining the diversity of the induced topics.
Speech-to-Speech Translation with Discrete-Unit-Based Style Transfer (2024.acl-srw)

Copied to clipboard

Challenge: Existing methods to translate spoken utterances from one language to another are unable to preserve speaker timbre of source speech.
Approach: They propose a pipeline with style-transfer capability on the basis of self-supervised speech representations and codec units.
Outcome: The proposed model achieves zero-shot cross-lingual style transfer on previously unseen source languages.
BiasDPO: Mitigating Bias in Language Models through Direct Preference Optimization (2024.acl-srw)

Copied to clipboard

Challenge: Large Language Models (LLMs) have been shown to be effective in complex language tasks, but their potential to perpetuate biases poses significant concerns.
Approach: They propose a new framework employing Direct Preference Optimization to mitigate biases in LLMs.
Outcome: The proposed model outperforms the baseline model on almost all bias benchmarks and achieves better performance than open-source models.
Document Alignment based on Overlapping Fixed-Length Segments (2024.acl-srw)

Copied to clipboard

Challenge: Existing studies show that web crawling can be used to obtain large-scale parallel corpora for NLP tasks.
Approach: They propose a sentence-based segmentation method for document alignment . they compare it with a fixed-length segmentation technique to handle long-text encoding better.
Outcome: The proposed method improves document alignment and recall by 1% to 10% on a document alignment task for Japanese-English and French-English datasets.
ReMAG-KR: Retrieval and Medically Assisted Generation with Knowledge Reduction for Medical Question Answering (2024.acl-srw)

Copied to clipboard

Challenge: Large Language Models (LLMs) have significant potential for facilitating intelligent end-user applications in healthcare, but hallucinations remain an inherent problem with LLMs.
Approach: They propose a pipeline to retrieve and medicalally-augmented-generation with knowledge reduction using cross-encoder re-ranking strategies to reduce the knowledge base.
Outcome: The proposed pipeline reduces the knowledge base and improves inference time by 47%.
Demystifying Instruction Mixing for Fine-tuning Large Language Models (2024.acl-srw)

Copied to clipboard

Challenge: Instruction tuning is effective for aligning large language models with human instructions, but the procedure to optimizing the mixing of instruction datasets is still unclear.
Approach: They categorize instructions into three primary types: NLP downstream tasks, coding, and general chat.
Outcome: The proposed method improves performance of large language models (LLMs) but it is difficult to combine different instruction datasets to optimize overall performance.
Fine-Tuning ASR models for Very Low-Resource Languages: A Study on Mvskoke (2024.acl-srw)

Copied to clipboard

Challenge: Recent advances in multilingual models for automatic speech recognition (ASR) have been able to achieve a high accuracy for languages with extremely limited resources.
Approach: They examine the parameter efficiency of training an adapter for the Mvskoke language, an indigenous language of America.
Outcome: The proposed model is parameter efficient and gives higher accuracy for a relatively small amount of data.
Automating Qualitative Data Analysis with Large Language Models (2024.acl-srw)

Copied to clipboard

Challenge: Existing methods for qualitative data analysis are far from resembling a human's analysis outcome.
Approach: They propose a method based on Large Language Models to tackle automated coding and make it as close as possible to the results of human researchers.
Outcome: The proposed method is based on large language models and can be as close as possible to the results of human researchers.
ANHALTEN: Cross-Lingual Transfer for German Token-Level Reference-Free Hallucination Detection (2024.acl-srw)

Copied to clipboard

Challenge: ANHALTEN is a new evaluation dataset that extends the English hallucination detection dataset to German.
Approach: They propose a dataset that extends the English hallucination detection dataset to German . they show that larger context length leads to better halluciation detection in german .
Outcome: ANHALTEN is the first evaluation dataset that extends the English hallucination detection dataset to German.
Label-Aware Automatic Verbalizer for Few-Shot Text Classification in Mid-To-Low Resource Languages (2024.acl-srw)

Copied to clipboard

Challenge: Prompt-based learning has shown its effectiveness in few-shot text classification.
Approach: They propose a prompt-based learning verbalizer that automatically selects a word to represent each class . they use the label name along with the conjunction "and" to induce the model to generate more effective words for the verbaliser.
Outcome: The proposed method outperforms existing verbalizers on four Southeast Asian languages.
Vector Spaces for Quantifying Disparity of Multiword Expressions in Annotated Text (2024.acl-srw)

Copied to clipboard

Challenge: We show that multiword expressions are a good study for linguistic diversity due to theiridiosyncratic nature.
Approach: They train static MWE-aware word embeddings for verbal MWEs in 14 languages . they find that the disparity measure aggregatingthem at a global scale correlates with the number of types .
Outcome: The proposed method is based on a set of vector spaces for VMWEs in 14 languages.
Narratives at Conflict: Computational Analysis of News Framing in Multilingual Disinformation Campaigns (2024.acl-srw)

Copied to clipboard

Challenge: Existing methods for multilingual framing differ from those used in English-speaking world . framers often use loaded vocabularies to create political images or favor a particular point of view .
Approach: They use eight years of Russian-backed disinformation campaigns to examine framing . they find that disinformation campaign consistently favors specific framers .
Outcome: The proposed method underperforms and shows high disagreements in Russian-language articles . the proposed method is based on eight years of Russian-backed disinformation campaigns .
Assessing In-context Learning and Fine-tuning for Topic Classification of German Web Data (2024.acl-srw)

Copied to clipboard

Challenge: Using a few hundred annotated data points per topic, we detect content related to three German policies in a database of scraped webpages.
Approach: They propose to use annotated data to train a binary classification task to detect topic-related content in a scraped database of webpages.
Outcome: The proposed model detects content related to three German policies in a scraped database of scrapes of webpages using a few hundred annotated data points per topic.
Knowledge Editing of Large Language Models Unconstrained by Word Order (2024.acl-srw)

Copied to clipboard

Challenge: Existing methods for identifying knowledge neurons for large language models have been challenging for black-boxed models . a new method is proposed to edit the knowledge held by the LLMs .
Approach: They propose a method that identifies the knowledge neurons that encode the target knowledge and adjusts the parameters associated with these neurons to update the knowledge.
Outcome: The proposed method outperforms existing methods on English and Japanese . it eliminates word order constraints and allows flexible locating regardless of the language .
Exploring the Effectiveness and Consistency of Task Selection in Intermediate-Task Transfer Learning (2024.acl-srw)

Copied to clipboard

Challenge: Identifying beneficial tasks to transfer from is a critical step toward successful intermediate-task transfer learning.
Approach: They propose a method that measures pairwise token similarity using maximum inner product search to improve task prediction.
Outcome: The proposed method improves task prediction scores from 2.59% to 3.96% for tasks requiring reasoning abilities, but not for reasoning abilities.
Does the structure of textual content have an impact on language models for automatic summarization? (2024.acl-srw)

Copied to clipboard

Challenge: Existing models for automatic summarization of long sequences suffer from context limitation.
Approach: They propose to take into account textual information coming from distinct passages from the long texts to be summarized.
Outcome: The proposed model improves on the performance of LongFormer on English.
Action Inference for Destination Prediction in Vision-and-Language Navigation (2024.acl-srw)

Copied to clipboard

Challenge: Existing work on vision-and-language navigation focuses on spatial reasoning and semantic grounding of visual information, but there is still scope for improvement.
Approach: They propose a VLN task of destination prediction for picking up a pedestrian that requires action inference from a crowd-sourced dataset.
Outcome: The proposed model can reason about the effect of the next action and the next on the destination to a certain extent.
A Computational Analysis and Exploration of Linguistic Borrowings in French Rap Lyrics (2024.acl-srw)

Copied to clipboard

Challenge: rap is a popular genre in the u.s. and has been used in countries far beyond the uk . linguistic borrowings are especially intriguing in countries such as the eu and europe .
Approach: They manually annotate a lexicon of over 700 borrowings in the French language . they find that there are increases in the proportion of linguistic borrowings, interjections, and Niger-Congo borrowings .
Outcome: The proposed method analyzes a corpus of over 8000 french rap song lyrics and shows that rap borrowings are increasing in prevalence and interjections are decreasing.
On Improving Repository-Level Code QA for Large Language Models (2024.acl-srw)

Copied to clipboard

Challenge: Commercial AI-assisted programming Chatbots may generate incorrect information when requests go beyond the model training data or require additional knowledge.
Approach: They propose to implement different self-alignment processes and retrieval-augmented generation pipelines to improve the copilot performance.
Outcome: The proposed model improves the copilot performance on repository-level semantics, dependency between files, and meta-information about the repository.
Compromesso! Italian Many-Shot Jailbreaks undermine the safety of Large Language Models (2024.acl-srw)

Copied to clipboard

Challenge: Despite efforts to align large language models with ethical guidelines, models can still be induced into unsafe behavior with jailbreaking.
Approach: They investigate the impact of many-shot jailbreaking on LLMs in italian . they find models exhibit unsafe behaviors even with minimal exposure to harmful prompts .
Outcome: The proposed model exhibits unsafe behaviors even with minimal exposure to harmful prompts, and this tendency rapidly escalates with more demonstrations.
ViMedAQA: A Vietnamese Medical Abstractive Question-Answering Dataset and Findings of Large Language Model (2024.acl-srw)

Copied to clipboard

Challenge: Existing abstractive question-answering datasets in Vietnamese are lacking .
Approach: They propose to introduce a Vietnamese abstractive question-answering corpus to address this gap . they propose to use Vietnamese abstractives to generate answers to questions .
Outcome: The proposed dataset examines the capability of large language models in the Vietnamese medical domain, including reasoning, memorizing and awareness of essential information.
Rescue: Ranking LLM Responses with Partial Ordering to Improve Response Generation (2024.acl-srw)

Copied to clipboard

Challenge: Customizing LLMs for a specific task involves separating high-quality responses from lower-quality ones. Obtaining a large volume of expert-annotated data is costly for most tasks.
Approach: They propose a method that trains the model to prioritize the best responses from a pool of candidates created for a task using ranking metrics.
Outcome: The proposed method is more robust, less sensitive to noise, and can be achieved with limited human annotations or through heuristic methods.
Basreh or Basra? Geoparsing Historical Locations in the Svoboda Diaries (2024.acl-srw)

Copied to clipboard

Challenge: In the historical domain, many geoparsing corpora are from large news collections.
Approach: They propose a pipeline employing named entity recognition for geotagging and a map-based generate-and-rank approach incorporating candidate name augmentation and clustering of location context words for geocoding.
Outcome: The proposed pipeline outperforms existing map-based geoparsers in terms of accuracy, lowest mean distance error, and number of locations correctly identified.
Homophone2Vec: Embedding Space Analysis for Empirical Evaluation of Phonological and Semantic Similarity (2024.acl-srw)

Copied to clipboard

Challenge: Existing studies have shown that homophones with different semantic/syntactic contexts are easier for children to memorize.
Approach: They propose a method for empirically evaluating the relationship between phonological and semantic similarity of linguistic units using embedding spaces.
Outcome: The proposed method shows that Chinese character homophones have a positive semantic relationship at varying levels of sound-sharing.
Trace-of-Thought Prompting: Investigating Prompt-Based Knowledge Distillation Through Question Decomposition (2024.acl-srw)

Copied to clipboard

Challenge: Trace-of-Thought Prompting allows small neural networks to emulate larger, teacher models with reduced computational demands.
Approach: They propose a framework to distill critical reasoning capabilities from teacher models to student models . they use problem decomposition to enhance interpretability and facilitate human-in-the-loop interventions .
Outcome: a new framework enables small neural networks to emulate the performance of larger, teacher models . it leverages problem decomposition to enhance interpretability and facilitate human-in-the-loop interventions . the proposed framework is available on github.com/trace-of-thought/trac-of_thought-prompting/main .
Can LLMs Augment Low-Resource Reading Comprehension Datasets? Opportunities and Challenges (2024.acl-srw)

Copied to clipboard

Challenge: Large Language Models (LLMs) have demonstrated impressive zero-shot performance on a wide range of NLP tasks.
Approach: They propose to use large language models to augment extractive reading comprehension datasets by fine-tuning their annotations and comparing their performance to human annotators.
Outcome: The proposed model can be used to augment extractive reading comprehension datasets.
Automatic Derivation of Semantic Representations for Thai Serial Verb Constructions: A Grammar-Based Approach (2024.acl-srw)

Copied to clipboard

Challenge: Using rich semantic representations for Thai Serial Verb Constructions (SVCs) is time-consuming and manual annotation is preferred.
Approach: They propose to implement an HPSG analysis for Thai Serial Verb Constructions (SVCs) they use a DELPH-IN computational grammar to generate appropriate representations from syntactic features.
Outcome: The proposed grammar increases verified coverage of Thai SVCs by 73% and decreases ambiguity by 46% on held-out data.
Bridging Distribution Gap via Semantic Rewriting with LLMs to Enhance OOD Robustness (2024.acl-srw)

Copied to clipboard

Challenge: Existing methods for fine-tuning on indistribution data fail to provide robustness against distribution shifts limiting the practical deployment of LLMs in dynamic real-world scenarios.
Approach: They propose a method that leverages the flexibility of LLMs to align both in-distribution (ID) and OOD data with the LLM's distributions.
Outcome: The proposed method outperforms fine-tuning methods on OOD tasks and benchmark datasets.
CoVoSwitch: Machine Translation of Synthetic Code-Switched Text Based on Intonation Units (2024.acl-srw)

Copied to clipboard

Challenge: Multilingual code-switching research is often hindered by the lack and linguistically biased status of available datasets.
Approach: They synthesize code-switching data by replacing intonation units detected through PSST, a speech segmentation model fine-tuned from OpenAI’s Whisper, using a language-to-text translation dataset, CoVoST 2.
Outcome: The proposed model outperforms two monolingual models and is better at code-switching translation into English than non-English.
Beyond Abstracts: A New Dataset, Prompt Design Strategy and Method for Biomedical Synthesis Generation (2024.acl-srw)

Copied to clipboard

Challenge: Existing methods to automate systematic reviews of papers are slow and incomplete . authors propose a new method to automating the systematic review process .
Approach: They propose a method for automatic synthesis generation using a dataset and prompting-based method.
Outcome: The proposed method improves the existing model and prompts the system to generate high-quality syntheses.
Improving Sentence Embeddings with Automatic Generation of Training Data Using Few-shot Examples (2024.acl-srw)

Copied to clipboard

Challenge: Decoder-based large language models (LLMs) have shown high performance on many tasks in natural language processing.
Approach: They propose to automatically generate an NLI dataset with an LLM and use it for fine-tuning of PromptEOL.
Outcome: The proposed model outperforms existing models on STS tasks without large manually annotated datasets.
Curriculum Learning for Small Code Language Models (2024.acl-srw)

Copied to clipboard

Challenge: Existing studies have shown that curriculum learning improves performance of code language models when it comes to complex tasks.
Approach: They propose a novel code difficulty assessment metric and introduce a Novel Curriculum Learning schedule that enhances the performance of small decoder-only language models.
Outcome: The proposed model improves on code execution tasks while its effect on code completion is less significant.
Question-Analysis Prompting Improves LLM Performance in Reasoning Tasks (2024.acl-srw)

Copied to clipboard

Challenge: Existing methods to improve LLM performance have focused on sophisticating the model's step-by-step calculation.
Approach: They propose a question analysis prompting strategy in which the model is prompted to explain the question in 'n' words before solving.
Outcome: The proposed prompt outperforms state-of-the-art prompts on arithmetic and commonsense datasets and consistently ranks among the top-2 prompts.
CheckersGPT: Learning World Models through Language Modeling (2024.acl-srw)

Copied to clipboard

Challenge: Large Language Models (LLMs) have shown impressive performance on various tasks, but the underlying process behind predicting the desired next token remains a black box.
Approach: They train a GPT-style autoregressive language model using only the next character prediction objective and then train corresponding model with different layer sizes.
Outcome: The proposed model shows a hint of learning a world model representation of the board positions on a simulated game of checkers and human gameplay dataset.
In-Context Symbolic Regression: Leveraging Large Language Models for Function Discovery (2024.acl-srw)

Copied to clipboard

Challenge: State of the art Symbolic Regression (SR) methods build specialized models, while the application of Large Language Models (LLMs) remains largely unexplored.
Approach: They propose a framework which iteratively refines a functional form with an LLM and determines its coefficients with an external optimizer.
Outcome: The proposed method outperforms the best SR methods on four popular benchmarks while yielding simpler equations with better out of distribution generalization.
STEP: Staged Parameter-Efficient Pre-training for Large Language Models (2024.acl-srw)

Copied to clipboard

Challenge: Existing methods for reducing computational costs during pre-training have been studied, but they often degrade performance under fair conditions.
Approach: They propose a method that combines parameter-efficient tuning and staged training to reduce memory requirements while maintaining comparable performance.
Outcome: The proposed method reduces memory requirements by 40.4% while maintaining comparable performance.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations