# An Empirical Study of Many-to-Many Summarization with Large Language Models

arXiv: https://arxiv.org/abs/2505.12983 · HTML: https://mlsanity.com/paper/2505.12983/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 33 of its 60 references that are in the map.
- First posted on arXiv 19 May 2025.

## Builds on (most cited first, 30 of 33)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [BERTScore: Evaluating Text Generation with BERT](https://mlsanity.com/paper/1904.09675/) (2019)
- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [Qwen Technical Report](https://mlsanity.com/paper/2309.16609/) (2023)
- [Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?](https://mlsanity.com/paper/2202.12837/) (2022)
- [Multilingual Denoising Pre-training for Neural Machine Translation](https://mlsanity.com/paper/2001.08210/) (2020)
- [A Survey on In-context Learning](https://mlsanity.com/paper/2301.00234/) (2022)
- [Baichuan 2: Open Large-scale Language Models](https://mlsanity.com/paper/2309.10305/) (2023)
- [The FLORES-101 Evaluation Benchmark for Low-Resource and Multilingual Machine Translation](https://mlsanity.com/paper/2106.03193/) (2021)
- [Is ChatGPT a General-Purpose Natural Language Processing Task Solver?](https://mlsanity.com/paper/2302.06476/) (2023)
- [Is ChatGPT a Good NLG Evaluator? A Preliminary Study](https://mlsanity.com/paper/2303.04048/) (2023)
- [XL-Sum: Large-Scale Multilingual Abstractive Summarization for 44 Languages](https://mlsanity.com/paper/2106.13822/) (2021)
- [Newsroom: A Dataset of 1.3 Million Summaries with Diverse Extractive Strategies](https://mlsanity.com/paper/1804.11283/) (2018)
- [WikiLingua: A New Benchmark Dataset for Cross-Lingual Abstractive Summarization](https://mlsanity.com/paper/2010.03093/) (2020)
- [Benchmark Data Contamination of Large Language Models: A Survey](https://mlsanity.com/paper/2406.04244/) (2024)
- [MLSUM: The Multilingual Summarization Corpus](https://mlsanity.com/paper/2004.14900/) (2020)
- [NCLS: Neural Cross-Lingual Summarization](https://mlsanity.com/paper/1909.00156/) (2019)
- [Detecting and Mitigating Hallucinations in Multilingual Summarisation](https://mlsanity.com/paper/2305.13632/) (2023)
- [mFACE: Multilingual Summarization with Factual Consistency Evaluation](https://mlsanity.com/paper/2212.10622/) (2022)
- [MT6: Multilingual Pretrained Text-to-Text Transformer with Translation Pairs](https://mlsanity.com/paper/2104.08692/) (2021)
- [A Survey on Cross-Lingual Summarization](https://mlsanity.com/paper/2203.12515/) (2022)
- [Models and Datasets for Cross-Lingual Summarisation](https://mlsanity.com/paper/2202.09583/) (2022)
- [Mixed-Lingual Pre-training for Cross-lingual Summarization](https://mlsanity.com/paper/2010.08892/) (2020)
- [A Variational Hierarchical Model for Neural Cross-Lingual Summarization](https://mlsanity.com/paper/2203.03820/) (2022)
- [Towards Unifying Multi-Lingual and Cross-Lingual Summarization](https://mlsanity.com/paper/2305.09220/) (2023)

## Abstract (arXiv)

Many-to-many summarization (M2MS) aims to process documents in any language and generate the corresponding summaries also in any language. Recently, large language models (LLMs) have shown strong multi-lingual abilities, giving them the potential to perform M2MS in real applications. This work presents a systematic empirical study on LLMs' M2MS ability. Specifically, we first reorganize M2MS data based on eight previous domain-specific datasets. The reorganized data contains 47.8K samples spanning five domains and six languages, which could be used to train and evaluate LLMs. Then, we benchmark 18 LLMs in a zero-shot manner and an instruction-tuning manner. Fine-tuned traditional models (e.g., mBART) are also conducted for comparisons. Our experiments reveal that, zero-shot LLMs achieve competitive results with fine-tuned traditional models. After instruct-tuning, open-source LLMs can significantly improve their M2MS ability, and outperform zero-shot LLMs (including GPT-4) in terms of automatic evaluations. In addition, we demonstrate that this task-specific improvement does not sacrifice the LLMs' general task-solving abilities. However, as revealed by our human evaluation, LLMs still face the factuality issue, and the instruction tuning might intensify the issue. Thus, how to control factual errors becomes the key when building LLM summarizers in real applications, and is worth noting in future research.