CADS: A Systematic Literature Review on the Challenges of Abstractive Dialogue Summarization
Authors: Frederic Kirstein, Jan Philip Wahle, Bela Gipp, Terry Ruas
JAIR 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Theoretical | This article summarizes the research on Transformer-based abstractive summarization for English dialogues by systematically reviewing 1262 unique research papers published between 2019 and 2024, relying on the Semantic Scholar and DBLP databases. |
| Researcher Affiliation | Academia | Frederic Kirstein EMAIL Jan Philip Wahle EMAIL Bela Gipp EMAIL Terry Ruas EMAIL Georg-August University Göttingen, Papendiek 14 Göttingen, 37073, Germany |
| Pseudocode | No | The paper describes methodologies and approaches for a systematic literature review in prose and tables (e.g., Table 1), but it does not include any explicitly labeled 'Pseudocode' or 'Algorithm' blocks, nor does it present structured, code-like steps for a method or procedure. |
| Open Source Code | Yes | All resources for our review are publicly available.1 1. https://github.com/FKIRSTE/Lit Rev-Dialogue Sum |
| Open Datasets | Yes | All resources for our review are publicly available.1 1. https://github.com/FKIRSTE/Lit Rev-Dialogue Sum |
| Dataset Splits | No | The paper describes a systematic literature review process, identifying 1262 unique papers and filtering them down to 133 for detailed analysis based on specific eligibility criteria. This process involves selection and exclusion, but it does not specify traditional training/test/validation dataset splits as would be used for machine learning experiments, because the paper itself is a review and does not train or evaluate a model. |
| Hardware Specification | No | The paper is a systematic literature review and focuses on analyzing existing research. It describes its methodology for selecting and categorizing papers, but it does not conduct its own computational experiments that would require specific hardware specifications for reproduction. |
| Software Dependencies | No | The paper mentions 'Employing a Python-based pipeline (Section 9)' for document extraction, but it does not provide specific version numbers for Python or any associated libraries or software dependencies used in their systematic review process. |
| Experiment Setup | No | The paper describes a systematic literature review and does not involve computational experiments with machine learning models. Therefore, it does not provide details on experimental setup, hyperparameters, or training configurations. |