Energy-Based Reasoners: 385x more efficient inference
discuss
- Points292
- Comments49
discuss
Story details from Hacker News ↗ · linked article on energy-based-reasoners.github.io.
Energy-Based Reasoners: 385x more efficient inference was first seen on Hacker News · Front page on Nov 18, 2026. Since then it has appeared in 63 checks on 2 different days. Its best position was #1 of 30, reached on Nov 18, 11:43 UTC. It dropped off the list after Nov 19, 04:43 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.
Recommend checking out Logical Intelligence, chaired by among others Yann LeCun, for more context on EBMs [0]. Has great potential in areas that LLMs traditionally do not perform that well in and, because they want to show the advantages of EBMs, also tends to be a great source of examples LLMs struggle with, like…
A clarifying way to think about this for sudoku: For every clue, mark the probability of sampling its number in the row/col to zero. Also mark the probability of selecting a "clue cell" for modification to 0. Pick a random cell for modification, based on its "energy" (higher energy get modified first): 1. Every…
I've spent a year on this. EBMs while ... how to say it, they are impressive validators , it's insanely hard to train them. They are very niche specific. While LLMs , being not right all the time and clumsy and so on, they help the people to get the job done at the end. EBM sudoku is not that hard in a small scale,…
Archived daily, weekly and monthly lists that include this entry.
Not at the moment. It was last on the list on Nov 19, 04:43 UTC, after 2 days of trending.
#1 on Hacker News · Front page, on Nov 18, 11:43 UTC. Trendoday checks this list every 15 minutes.
It was first seen on Nov 18, 2026 and has been on the list on 2 days (63 checks) since.
292 points and 49 comments; submitted by frenchmajesty on Nov 18, 2026.