Prithviraj Ammanabrolu, UC San Diego assistant professor and Nvidia research scientist, introduces Introspective X Training to consolidate LLM pre-training, mid-training, and post-training into one offline RL-inspired process

Ever wished we had fewer X-training hyphenates? Pre, mid, post etc. Why not just Training?

Trying to bridge the divides (and get all our friends into one team again), we intro *Introspective X Training*, an offline RL inspired method that scales effectively across any LLM stage by annotating your data with a thinking reward generated language critique!

Up to 2.8x FLOP efficiency + 5-10 point score gains (esp with math and code) at any stage from scratch to 24T tokens on 8b (active) sized models!! We burned much compute ablating so you wouldn't have to

Moral of the story is‼️don't throw out any data via filtering, just feedback condition it‼️

You can spend FLOPs up front on inference to *classify* data quality and then train so that tokens aren't all treated equally based on the feedback starting early in training itself. Right now they're really only separated out much later during mid/post training

This improves overall compute efficiency and gives us benchmark perf not possible with just baseline methods!

Paper here: https://arxiv.org/abs/2605.20285 Thanks to @BrandoCui and @GXiming for leading this w/ @__SyedaAkter @davidjesusacu @hyunw_kim @jaehunjung_com Yuxiao Qu @shrimai_ @YejinChoinka