FABLES: Evaluating faithfulness and content selection in book-length summarization ยท HackerPapers discussionshttps://hackerpaper.kaipfister.de/p/2404.01261/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMTFABLES: Evaluating faithfulness and content selection in book-length summarization โ€” mentioned by nsagenthttps://news.ycombinator.com/item?id=40360910urn:hackerpapers:mention:2404.01261:40360910Tue, 14 May 2024 22:31:52 GMT<p>A number of people in my lab do research into long context evaluation of LLMs for works of fiction. The likelihood is very high that Moby Dick is in the training data. Instead the people in my lab have explored recently published books to avoid these issues. See BooookScore (https://openreview.net/forum?id=7Ttk3RzDeu) which was just presented at ICLR last week and FABLES (https://arxiv.org/abs/2404.01261) a recent preprint.</p><p><a href="https://hackerpaper.kaipfister.de/p/2404.01261/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.CLFABLES: Evaluating faithfulness and content selection in book-length summarization โ€” HN submissionhttps://news.ycombinator.com/item?id=39982362urn:hackerpapers:submission:2404.01261:39982362Tue, 09 Apr 2024 18:14:47 GMT<p>Evaluating faithfulness and content selection of LLMs in book-length summaries</p><p></p><p>71 points; 7 discussion replies on HN. <a href="https://hackerpaper.kaipfister.de/p/2404.01261/">Paper and source history</a>. A submission is not an endorsement.</p>cs.CLFABLES: Evaluating faithfulness and content selection in book-length summarization โ€” HN submissionhttps://news.ycombinator.com/item?id=39957207urn:hackerpapers:submission:2404.01261:39957207Sun, 07 Apr 2024 01:03:50 GMT<p>Faithfulness and content selection in book-length summarization</p><p></p><p>2 points; 0 discussion replies on HN. <a href="https://hackerpaper.kaipfister.de/p/2404.01261/">Paper and source history</a>. A submission is not an endorsement.</p>cs.CL