Llamax: Scaling linguistic horizons of llm by enhancing translation capabilities beyond 100 languages

Y Lu, W Zhu, L Li, Y Qiao, F Yuan - arxiv preprint arxiv:2407.05975, 2024 - arxiv.org
Large Language Models (LLMs) demonstrate remarkable translation capabilities in high-
resource language tasks, yet their performance in low-resource languages is hindered by …

Scaling of search and learning: A roadmap to reproduce o1 from reinforcement learning perspective

Z Zeng, Q Cheng, Z Yin, B Wang, S Li, Y Zhou… - arxiv preprint arxiv …, 2024 - arxiv.org
OpenAI o1 represents a significant milestone in Artificial Inteiligence, which achieves expert-
level performances on many challanging tasks that require strong reasoning ability. OpenAI …

Fairness definitions in language models explained

TV Doan, Z Chu, Z Wang, W Zhang - arxiv preprint arxiv:2407.18454, 2024 - arxiv.org
Language Models (LMs) have demonstrated exceptional performance across various
Natural Language Processing (NLP) tasks. Despite these advancements, LMs can inherit …

Lumberchunker: Long-form narrative document segmentation

AV Duarte, J Marques, M Graça, M Freire, L Li… - arxiv preprint arxiv …, 2024 - arxiv.org
Modern NLP tasks increasingly rely on dense retrieval methods to access up-to-date and
relevant contextual information. We are motivated by the premise that retrieval benefits from …

xTower: A multilingual LLM for explaining and correcting translation errors

M Treviso, NM Guerreiro, S Agrawal, R Rei… - arxiv preprint arxiv …, 2024 - arxiv.org
While machine translation (MT) systems are achieving increasingly strong performance on
benchmarks, they often produce translations with errors and anomalies. Understanding …

Emma-500: Enhancing massively multilingual adaptation of large language models

S Ji, Z Li, I Paul, J Paavola, P Lin, P Chen… - arxiv preprint arxiv …, 2024 - arxiv.org
In this work, we introduce EMMA-500, a large-scale multilingual language model continue-
trained on texts across 546 languages designed for enhanced multilingual performance …

Beyond English-Centric LLMs: What Language Do Multilingual Language Models Think in?

C Zhong, F Cheng, Q Liu, J Jiang, Z Wan… - arxiv preprint arxiv …, 2024 - arxiv.org
In this study, we investigate whether non-English-centric LLMs, despite their strong
performance,think'in their respective dominant language: more precisely,think'refers to how …

DiBiMT: A Gold Evaluation Benchmark for Studying Lexical Ambiguity in Machine Translation

F Martelli, S Perrella, N Campolungo… - Computational …, 2024 - direct.mit.edu
Despite the remarkable progress made in the field of Machine Translation (MT), current
systems still struggle when translating ambiguous words, especially when these express …

Enhancing translation accuracy of large language models through continual pre-training on parallel data

M Kondo, T Utsuro, M Nagata - arxiv preprint arxiv:2407.03145, 2024 - arxiv.org
In this paper, we propose a two-phase training approach where pre-trained large language
models are continually pre-trained on parallel data and then supervised fine-tuned with a …

Prexme! large scale prompt exploration of open source llms for machine translation and summarization evaluation

C Leiter, S Eger - arxiv preprint arxiv:2406.18528, 2024 - arxiv.org
Large language models (LLMs) have revolutionized the field of NLP. Notably, their in-
context learning capabilities also enable their use as evaluation metrics for natural language …