Previous-Token Prediction Based LLM Inversion for Prompt Reconstruction
discuss
- Points4
- Comments1
discuss
Story details from Hacker News ↗ · linked article on arxiv.org.
Previous-Token Prediction Based LLM Inversion for Prompt Reconstruction was first seen on Hacker News · Front page on Nov 19, 2026. Since then it has appeared in 22 checks on 1 different day. Its best position was #11 of 30, reached on Nov 19, 03:43 UTC. It dropped off the list after Nov 19, 07:14 UTC.
Position on Hacker News · Front page at every check (lower is better).
Best rank per day (UTC).
Platform metric as reported by Hacker News.
A few top-level comments from Hacker News, shortened. Read the full thread on Hacker News.
Imagine if you could take the AI marketing slop, or political smear campaigns, texts, and Images, and see the misanthropic sludge that was used as a prompt to generate it.
Archived daily, weekly and monthly lists that include this entry.
Not at the moment. It was last on the list on Nov 19, 07:14 UTC, after 1 day of trending.
#11 on Hacker News · Front page, on Nov 19, 03:43 UTC. Trendoday checks this list every 15 minutes.
It was first seen on Nov 19, 2026 and has been on the list on 1 day (22 checks) since.
4 points and 1 comment; submitted by dalmo3 on Nov 19, 2026.