Papers by Alham Aji

15 papers
M4: Multi-generator, Multi-domain, and Multi-lingual Black-Box Machine-Generated Text Detection (2024.eacl-long)

Copied to clipboard

Challenge: Large language models generate fluent responses to user queries, but they are also susceptible to misuse in journalism, education, and academia.
Approach: They propose a large-scale benchmark for machine-generated text detection that is a multi-generator, multi-domain, and multi-lingual corpus.
Outcome: The proposed system can detect machine-generated text and pinpoint misuse . the proposed system is based on a large-scale benchmark dataset .
Re-Evaluating Evaluation for Multilingual Summarization (2024.emnlp-main)

Copied to clipboard

Challenge: Existing studies have shown that automated evaluation approaches correlate with human ratings in English, but this is unclear for other languages.
Approach: They construct a small-scale pilot dataset containing article-summary pairs and human ratings in English, Chinese and Indonesian to measure the strength of summaries.
Outcome: The results show that standard metrics are unreliable measures of quality in Chinese and Indonesian.
COPAL-ID: Indonesian Language Reasoning with Local Culture and Nuances (2024.naacl-long)

Copied to clipboard

Challenge: Existing multilingual language models struggle to capture local nuances and contexts that vary from culture to culture.
Approach: They propose a public Indonesian language common sense reasoning dataset COPAL-ID . it incorporates Indonesian local and cultural nuances and provides a more natural portrayal of causal reasoning .
Outcome: The proposed dataset is fluent and free from awkward phrases, unlike the previous dataset.
Efficient and Interpretable Grammatical Error Correction with Mixture of Experts (2024.findings-emnlp)

Copied to clipboard

Challenge: Error type information has been widely used to improve the performance of grammatical error correction models.
Approach: They propose a mixture-of-experts model for grammatical error correction that uses error type information to generate corrections and combine models.
Outcome: The proposed model achieves the performance of T5-XL with three times fewer effective parameters and produces interpretable corrections by also identifying the error type during inference.
M4GT-Bench: Evaluation Benchmark for Black-Box Machine-Generated Text Detection (2024.acl-long)

Copied to clipboard

Challenge: Large Language Models (LLMs) have brought an unprecedented surge in machine-generated text (MGT) societal implications are posed by their potential misuse and lack of training data.
Approach: They propose a benchmark to detect machine-generated text in multiple languages . they use multi-domain and multi-generator corpus to identify which model generated the text .
Outcome: The proposed benchmark compares a multilingual, multi-domain and multi-generator corpus of MGTs with human-generated content.
SemRel2024: A Collection of Semantic Textual Relatedness Datasets for 13 Languages (2024.findings-acl)

Copied to clipboard

Challenge: SemRel datasets are annotated by native speakers across 13 languages . they are used to characterise the relationship between two units of text .
Approach: They propose to use a semantic relatedness dataset to measure the degree of semantic textual relatedness between sentences in Afrikaans, Algerian Arabic, Amharic, English, Hausa, Hindi, Indonesian, Kinyarwanda, Marathi, Moroccan Arabic, Modern Standard Arabic, Spanish, and Telugu.
Outcome: The proposed datasets are annotated by native speakers across 13 languages and represent the semantic relatedness of 13 languages.
Towards Measuring and Modeling “Culture” in LLMs: A Survey (2024.emnlp-main)

Copied to clipboard

Challenge: Existing models are biased towards Western, Anglocentric or American cultures, a problem that is arguably detrimental to the performance of LLMs.
Approach: They analyze more than 90 recent papers that aim to study cultural representation and inclusion in large language models.
Outcome: The proposed models are biased towards Western, Anglocentric or American cultures, despite their diversity and their robustness.
Multilingual Large Language Models Are Not (Yet) Code-Switchers (2023.emnlp-main)

Copied to clipboard

Challenge: Existing multilingual Large Language Models are not specifically trained with objectives for managing code-switching scenarios.
Approach: They propose to use multilingual Large Language Models to perform sentiment analysis, machine translation, summarization and word-level language identification to compare their performance to fine-tuned models of much smaller scales.
Outcome: The proposed models show that they underperform in comparison to fine-tuned models of much smaller scales.
LLM-DetectAIve: a Tool for Fine-Grained Machine-Generated Text Detection (2024.emnlp-demo)

Copied to clipboard

Challenge: a large number of machine-generated texts are often hard to distinguish between human-written and machine-generated text . this raises concerns about potential misuse, especially within educational and academic domains .
Approach: They propose a system that can detect whether a text is human-written or machine-generated . they use a fine-grained classification schema to identify the use of machine-generated text .
Outcome: The proposed system can distinguish between human-written and machine-generated text . it can detect attempts to obfuscate the fact that a text was machine- generated .
GlobalBench: A Benchmark for Global Progress in Natural Language Processing (2023.emnlp-main)

Copied to clipboard

Challenge: despite advances in NLP, significant disparities in performance across languages still exist . prior benchmarks focused on a limited number of tasks and languages, but now GlobalBench tracks progress on all languages.
Approach: They propose to use global benchmarks to track progress on all NLP datasets in all languages.
Outcome: a new tool tracks progress on all NLP datasets in all languages and tracks per-speaker utility and equity . globalbench is designed to identify the most under-served languages and reward research efforts . a globalbech is available at https://github.com/neulab/globalbench.
LinguAlchemy: Fusing Typological and Geographical Elements for Unseen Language Generalization (2024.findings-emnlp)

Copied to clipboard

Challenge: Pretrained language models have shown remarkable generalization toward multiple tasks and languages, but their generalization towards unseen languages is poor.
Approach: They propose a regularization technique that incorporates various aspects of languages to better characterize linguistics constraints.
Outcome: The proposed technique improves accuracy of mBERT and XLM-R on unseen languages by 18% and 2% compared to fully finetuned models.
LaMini-LM: A Diverse Herd of Distilled Models from Large-Scale Instructions (2024.eacl-long)

Copied to clipboard

Challenge: Large language models with instruction tuning are resource-intensive . a recent study suggests that the performance of LLMs scales proportionally with the size of the model.
Approach: They propose to distill knowledge from instruction-tuned LLMs into much smaller ones . they develop a large set of 2.58M instructions based on existing and newly-generated instructions .
Outcome: The proposed models are comparable to strong baselines while being much smaller in size.
Cultural Conditioning or Placebo? On the Effectiveness of Socio-Demographic Prompting (2024.emnlp-main)

Copied to clipboard

Challenge: Socio-demographic prompting is a commonly employed approach to study cultural biases in LLMs as well as for aligning models to certain cultures.
Approach: They propose to use socio-demographic prompting to probe four LLMs with culturally sensitive and non-sensitive cues on datasets that are supposed to be culturally neutral or sensitive.
Outcome: The proposed model shows significant differences in responses on both kinds of datasets, casting doubt on its robustness.
LLM-powered Data Augmentation for Enhanced Cross-lingual Performance (2023.emnlp-main)

Copied to clipboard

Challenge: Existing training data for multilingual commonsense reasoning datasets is limited.
Approach: They propose to use large language models for data augmentation in multilingual datasets . they use Dolly-v2, StableVicuna, ChatGPT, and GPT-4 to augment three datasets.
Outcome: The proposed model outperforms larger general-purpose, zero-shot models when training in smaller models.
SEACrowd: A Multilingual Multimodal Data Hub and Benchmark Suite for Southeast Asian Languages (2024.emnlp-main)

Copied to clipboard

Challenge: Southeast Asia (SEA) is home to over 1,300 indigenous languages and 671 million people . prevailing AI models suffer from a significant lack of representation of texts, images, and audio datasets from SEA .
Approach: They propose to provide a resource center that provides standardized corpora in nearly 1,000 SEA languages across three modalities.
Outcome: a new benchmark assesses the quality of AI models on 36 SEA languages across 13 tasks . the results highlight the importance of SEA as a culturally diverse region .

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