Open Models Almanac
Methodspost-trainingpreference optimization

general family · filed under post-training

Reinforcement Learning from Human Feedback (RLHF)

A broad post-training approach that uses human feedback to guide reinforcement learning.

The sources on this page name it without explaining how it works. The line above is the extraction's one-line gloss. See preference optimization for the methods filed beside it.

Also called Reinforcement Learning from Human Feedback, RLHF.

sources
7
models
3
lab adopt it
1
strongest
core

Documented in

Evidence

8 quotes from 7 sources, one card per model, each opening with its strongest.

How sources treat it
used 7core 1

We train this teacher primarily using RLHF.

coreunclearin Nemotron 3.5 alignment teacherNVIDIA
4 more from this model

Super’s post-training pipeline combines supervised fine-tuning, Reinforcement Learning from Verifiable Rewards (RLVR), Reinforcement Learning from Human Feedback (RLHF)

usedpost trainingin Nemotron 3.5 SuperNVIDIA

Reinforcement learning with verifiable rewards, reinforcement learning from human feedback, and specialized teacher training

usedpost trainingin Nemotron 3.5 Super VLNVIDIA

Reinforcement Learning from Human Feedback (RLHF)

usedpost trainingin Nemotron 3.5 SuperNVIDIA

reinforcement learning from human feedback

usedpost trainingin Nemotron 3.5 Super VLNVIDIA

NVIDIA-Nemotron-3-Ultra-550B-A55B

NVIDIAadopted3 sources

Conversational quality was further refined through RLHF

usedoptimizationin Nemotron 3 UltraNVIDIA
2 more from this model

Chat teacher training involves multiple RLHF iterations

usedoptimizationin Nemotron 3 UltraNVIDIA

Nemotron 3 Ultra 550B-A55B GenRM: GenRM used for RLHF

usedpost trainingin Nemotron 3 UltraNVIDIA

Further reading

Picked by hand, not extracted: where to read more, not evidence for anything on this page.

↑