#artificial intelligence
Jun 2026
When Summaries Distort Decisions: Information Fidelity in LLM-Compressed Financial Analysis
Across financial filings and earnings-call transcripts, it is found that LLM-based compression can produce fluent and factually plausible compressed contexts that nevertheless alter downstream decisions, suggesting that financial compression should be evaluated not only by efficiency or factuality, but also by its abil...
Hoyoung Lee, Suhwan Park, Seung-Han Lee et al.
· arXiv.org · 3 citations