Cross-Lingual Summarization for Overseas Applications Using Multilingual Pre-Trained Models and Knowledge Distillation
Bibliographic record
Abstract
Cross-lingual summarization (CLS) has emerged as an essential tool for bridging linguistic gaps in global communication, facilitating the extraction of concise and meaningful content in a target language from source documents written in different languages. This work introduces a novel framework for CLS, leveraging state-of-the-art multilingual pre-trained models alongside knowledge distillation techniques to ensure efficient and scalable performance. The approach is built on transformer-based architectures, such as mBART and mT5, which are fine-tuned to address the challenges of semantic coherence, fluency, and context preservation across diverse language pairs. To optimize these models for deployment in resource-constrained environments, knowledge distillation is employed, transferring knowledge from larger, computationally expensive models to lightweight student models while maintaining high accuracy. Extensive experimentation on multilingual datasets showcases the robustness of the proposed method, achieving significant improvements over traditional pipeline approaches that rely on separate translation and summarization steps. Evaluation metrics, including ROUGE and BLEU scores, demonstrate superior performance across multiple language pairs. Moreover, the proposed framework achieves a threefold reduction in computational overhead compared to full-scale transformer models, making it highly suitable for real-world overseas applications. This research provides a scalable, efficient, and effective solution for generating high-quality summaries across languages, contributing to enhanced multilingual accessibility and information sharing on a global scale.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.006 |
| Meta-epidemiology (narrow) | 0.003 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.004 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.005 | 0.005 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".