{"id":1684,"date":"2025-10-17T08:08:31","date_gmt":"2025-10-17T08:08:31","guid":{"rendered":"https:\/\/prg.inf.unibe.ch\/?page_id=1684"},"modified":"2025-10-17T08:08:31","modified_gmt":"2025-10-17T08:08:31","slug":"thesis-scope","status":"publish","type":"page","link":"https:\/\/prg.inf.unibe.ch\/index.php\/education\/thesis-scope\/","title":{"rendered":"thesis-SCOPE"},"content":{"rendered":"\n
<\/div>\n\n\n\n
<\/div>
\n
<\/div>\n

SCOPE: Summarization with Context-Oriented Preprocessing and Evaluation<\/strong>\u00a0<\/h1>\n\n\n\n

Co-Supervised by:<\/strong> Merlin Streilein<\/p>\n\n\n\n

If you are interested in this topic or have further questions, do not hesitate to contact merlin.streilein@unibe.ch<\/a>.<\/p>\n\n\n\n

Background \/ Context<\/strong><\/p>\n\n\n\n

Modern text summarization models, such as BART or T5, are limited by their maximum input length (commonly around 1024 tokens). This presents a challenge for summarizing long documents such as scientific articles, legal contracts, or books. A naive solution is recursive summarization<\/em>, where documents are chunked and summarized iteratively, but this often leads to information loss, redundancy, or inconsistency. Recent research has suggested alternatives: heuristics that focus on specific parts of the document (e.g., beginnings and ends of paragraphs), importance-based ranking of sentences, or hybrid approaches combining extractive and abstractive methods. Despite these ideas, there has been little systematic evaluation of strategies to adapt fixed-length models to long inputs, and the potential for new preprocessing schemes remains\u00a0underexplored.\u00a0<\/p>\n\n\n\n

Research Question(s) \/ Goals<\/strong><\/p>\n\n\n\n