Check out our new paper! Really enjoyed working on this with @SachinKonan and @liuzhuang1234!
We fine-tune a model to predict conference acceptance and assess research quality, paving the way for evaluating automated research and measuring improvements in scientific writing.
Will your ICLR paper get accepted? ๐
Introducing PaperScorer, our NeurIPS '26 project that predicts acceptance from anonymized submissions.
It beats GPT, Gemini, and Jev, with well-calibrated probabilities.
Try it on your paper today: https://t.co/27XsDaHZzF
Blog: https://t.co/QxJkwCGA9i
Details in thread ๐งต
Will your ICLR paper get accepted? ๐
Introducing PaperScorer, our NeurIPS '26 project that predicts acceptance from anonymized submissions.
It beats GPT, Gemini, and Jev, with well-calibrated probabilities.
Try it on your paper today: https://t.co/27XsDaHZzF
Blog: https://t.co/QxJkwCGA9i
Details in thread ๐งต
@DanAdvantage Completed! :) Original NetHack code relied on globals to store state, forcing NLE to use dlopen for multi-process isolation. I refactored to strip those globals out, allowing us to parallelize via threads. Env was merged earlier today. What would you like to see re: state sets?
Presenting poster on generative diffusion models for DNA synthesis with @KiaGhods at ICLRโs Generative AI in Genomics workshop.
Come by Room 211 at 1pm if interested!
Paper: https://t.co/FNXTacUNCL