Hacker News Reader

It would've been interesting to compare models with larger context windows, e.g. Gemini with 1m+ tokens and Claude Opus. Otherwise, the title maybe should've been Financial Statement Analysis with GPT-4.

View on HN

The study only captured the financial statements. I am unsure what a larger context window would buy you.

You could dump all of a companies financial statements together, or dump all of a companies competitors in with it, for one.

That was outside the context of what the paper was studying.

maybe the authors should have used a larger context window in order to output their paper ;-)

I jest - a well contained and specified context is important for a paper.

But also, questions like "it would be interesting to also try this other thing" is how new papers get written.

Reply on news.ycombinator.com