general family · filed under post-training
Joint SFT and RL
A post-training approach described as combining supervised fine-tuning and reinforcement learning jointly, rather than as sequential stages.
- source
- 1
- model
- 1
- lab adopt it
- 1
- strongest
- used
How sources treat it
One count per evidence span, weakest treatment to strongest.
used 1
Documented in
Evidence
1 span quoted from the sources, strongest treatment first.
Joint SFT and RL improved coreference and constraint tracking.
usedtraining objectivein Hy3Tencent
Filed alongside
Other methods under post-training.
Scalable RL at Agent ScaleReinforcement learning for low-pass-rate tasksScalable reinforcement learning post-trainingSFT followed by RL and on-policy distillationExtended post-trainingPost-training on diverse domainsPost-training optimizationThree-stage post-training with multi-teacher on-policy distillationThree-stage Thinker post-training strategy