Twitxy Saved

Helen Qu

@_helenqu

research fellow @FlatironCCA working on multi-agent AI safety / AI for science. prev: PhD @physatpenn ‘24, BSE @CIS_Penn '17.
nyc helenqu.com Joined May 2012
292 Following 1.2K Followers 128 Posts

Latest posts 5

Helen Qu @_helenqu
+1! it’s hard to have a cohesive community working on alignment without having a shared working definition of what alignment really means
David Verified account @DavidDAfrica Nov 5
Quick new post: I think it'd be good if people (particularly newcomers) spent more time writing up existing takes from papers, and resolving contradictory evidence in their results and experiments. I think this is on average a better use of time and builds more understanding than firing off an experiment. And if you slop out both, this is cheaper to fail on. I then give a loose example on trying to resolve if reward hacking in RL generalises to emergent misalignment.

lesswrong.com/posts/tn4WSyrq…
View quoted post on X
Helen Qu @_helenqu
+1 to all of this! game theory/cooperation will be incredibly valuable to thinking about multi-agent alignment
Alexandra Barr Verified account @BarrAlexandra Oct 21
what should i read?

- canonical arxiv / blog posts, etc
- more philosophy? virtue ethics?
- economics? i feel bullish on econ theory teaching me something (collective action, game theory)
- what about psych? cooperation? honesty?
- neuroscience seems better for interp work vs HD
View quoted post on X