Booth 21-25 | AI Data Management Zone | Tokyo Big Sight
For most of the last six years, the most capable LLMs were locked behind APIs. You could call them, you could build on them, but you couldn’t look inside, customize
Before you reach for RLHF, before you design a reward model, before you start thinking about reinforcement learning from verifiable rewards — there’s a more fundamental question worth asking: has this model been properly fine-tuned on examples of the behavior you want? Supervised Fine-Tuning (SFT) sits between pre-training and the
