Line of inquiry
Inquiring lines›How should we train models for cap…›How do attention and architecture…›this line of inquiry

How do policy learning algorithm choices affect multi-objective optimization stability?

A broader line of inquiry — a family of 14 specific questions the research asks around this. Follow one into its inquiring-line page, or move sideways to a related line below.

Questions in this line of inquiry 14

Specific inquiring lines the field asks around this — ordered from the most general framing down to the most specific angle.

Related lines of inquiry 6

Neighbouring lines of inquiry by meaning — move sideways to an adjacent framing.

Can alternative training methods improve on supervised fine-tuning for language models?Can language model RL training avoid reward hacking and misalignment?What pretraining choices and baseline capability constrain reinforcement learning gains?What constrains reinforcement learning's ability to expand model reasoning?Why does reinforcement learning suppress output diversity compared to supervised fine-tuning?How can identical external performance mask different internal representations?