Papers › Multi-objective Reinforcement learning from AI Feedback

Multi-objective Reinforcement learning from AI Feedback

11 Jun 2024arXiv:2406.07295archive 2025-07-28

Marcus Williams

This paper presents Multi-Objective Reinforcement Learning from AI Feedback (MORLAIF), a novel approach to improving the alignment and performance of language models trained using reinforcement learning from AI feedback (RLAIF). In contrast to standard approaches that train a single preference model to represent all human preferences, MORLAIF decomposes this task into multiple simpler principles, such as toxicity, factuality, and sycophancy. Separate preference models are trained for each principle using feedback from GPT-3.5-Turbo. These preference model scores are then combined using different scalarization functions to provide a reward signal for Proximal Policy Optimization (PPO) training of the target language model. Our experiments indicate that MORLAIF outperforms the standard RLAIF baselines and that MORLAIF can be used to align larger language models using smaller ones. Surprisingly, the choice of scalarization function does not appear to significantly impact the results.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

carolius/Multi-Objective-Reinforcement-Learning-from-AI-Feedback officialmentioned in papermentioned on GitHubpytorchMIT report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Language ModelingLanguage ModellingMulti-Objective Reinforcement LearningReinforcement Learningreinforcement-learning

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

ALIGNAdamAttentionAttention DropoutBPECosine AnnealingDense ConnectionsDropoutGPT-3Layer NormalizationLinear LayerLinear Warmup With Cosine AnnealingMulti-Head AttentionRLAIFResidual ConnectionSoftmaxWeight Decay

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections