Autoresearch on an old research idea

Re:Autoresearch on an old research idea by Xx_crazy420_xX on Tuesday March 24, 2026 @12:55AM (#47498739)

Autoresearch is nothing new, big players are already in the game with more sophisticated solutions:

  - https://arxiv.org/abs/2602.02660 (MARS)
  - https://arxiv.org/abs/2601.14525 (Execution-grounded automated AI research)
  - https://arxiv.org/abs/2601.10402 (ML-Master 2.0)

The mostly used benchmark for automated AI engineering/ research is: https://github.com/openai/mle-bench

[ Parent | on HN ]

Re:Autoresearch on an old research idea by bluequbit on Tuesday March 24, 2026 @05:20AM (#47500222)

The thing is, autoresearch feels more accessible that the listed solutions. I can use it trivially on virtually any problem that has verifiable rewards and a feedback loop.

[ Parent | on HN ]

Re:Autoresearch on an old research idea by baxtr on Tuesday March 24, 2026 @05:39AM (#47500357)

People underestimate UX and accessibility. The iPhone was nothing new.

[ Parent | on HN ]

Re:Autoresearch on an old research idea by svnt on Tuesday March 24, 2026 @10:12AM (#47502903)

That’s because it is literally just a feedback loop?

[ Parent | on HN ]