paper + opportunity for current grad students

[Evan Miyazono]

Hi all,

## FM benchmarks

I recently flagged CLEVER from Thakur et al (from mostly UT Austin folks)

Feels worth highlighting Verina by Ye et al (from mostly Berkeley folks: Dawn's group + Kaiyu at FAIR)

I'd love to know of there are other papers related to GSAI that people should be tracking.

## Opportunity for grad students

If anyone has a grad student (finishing an MS or something like 3rd year) who is interested in coauthoring something with an established figure in the GSAI space, I'm happy to nominate. I suspect most people on this list were asked to nominate someone already, but on the off chance you were not, feel free to send students my way and I can review/pass along nominations.

On Mon, May 26, 2025 at 6:36 PM, Evan Miyazono <ev...@atlascomputing.org> wrote:

Hey GSAI community, 

I'd be surprised if all of this is new, but I'd also be surprised if none is:

Interesting benchmark for those who care about FM: CLEVER for end-to-end verified code generation in Lean: https://arxiv.org/abs/2505.13938

Also, some fellowship programs:

• Speculative Technologies is organizing a cohort to train potential ARPA PMs on AI-risk topics: https://spec.tech/ai-brains (part time, remote friendly, 3 months, apply by June 16)
  
• also, related to this, I found out there are some additional resources if you know someone who might be interested in being a DARPA PM.  (Please send me candidates!)
• Future of Life Foundation is organizing a fellowship to incubate new potential projects/orgs: https://www.flf.org/fellowship (full time, silicon valley, 12 weeks; apply by June 9)
  
• Constellation is running an incubator for new orgs https://constellation.org/programs/constellation-incubator (full time?, east bay, 4 months, rolling admission)

Best,

   - Evan