Objective Decoupling in Social Reinforcement Learning: Recovering Ground Truth from Sycophantic Majorities
On the map: Deep reinforcement learning methods → RL training of language and agent models
Abstract (arXiv)
Contemporary AI alignment strategies rely on a fragile premise: that human feedback, while noisy, remains a fundamentally truthful signal. In this paper, we identify this assumption as Dogma 4 of Reinforcement Learning (RL). We demonstrate that while this dogma holds in static environments, it fails in social settings where evaluators may be sycophantic, lazy, or adversarial. We prove that under Dogma 4, standard RL agents suffer from what we call Objective Decoupling, a structural failure mode where the agent's learned objective permanently separates from the latent ground truth, guaranteeing convergence to misalignment. To resolve this, we propose Epistemic Source Alignment (ESA). Unlike standard robust methods that rely on statistical consensus (trusting the majority), ESA utilizes sparse safety axioms to judge the source of the feedback rather than the signal itself. We prove that this "judging the judges" mechanism guarantees convergence to the true objective, even when a majority of evaluators are biased. Empirically, we show that while traditional consensus methods fail under majority collusion, our approach successfully recovers the optimal policy.
Key facts. Cited by 2 papers in the map (data through 13 Oct 2026). Placed in RL training of language and agent models, one of 15 directions in Deep reinforcement learning methods. Builds on 15 of its 31 references that are in the map. First posted on arXiv 8 Feb 2026.
Builds on
15 of 31 references.
Closest ancestors · 1
Same direction, newest first
Other references · 14
- 2022Training language models to follow instructions with human feedbackcited by 15,856
- 2017Deep reinforcement learning from human preferencescited by 3,871
- 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,691
- 2016Generative Adversarial Imitation Learningcited by 1,577
- 2017Curiosity-driven Exploration by Self-supervised Predictioncited by 1,180
- 2022Scaling Laws for Reward Model Overoptimizationcited by 1,077
- 2023Towards Understanding Sycophancy in Language Modelscited by 1,018
- 2024Deep Reinforcement Learning for Robotics: A Survey of Real-World Successescited by 155
- 2018Reinforcement Learning with Perturbed Rewardscited by 76
- 2024RRM: Robust Reward Model Training Mitigates Reward Hackingcited by 76
- 2015A Survey on Truth Discoverycited by 36
- 2025Robust LLM Alignment via Distributionally Robust Direct Preference Optimizationcited by 18
2 more
Built on by
Citing papers per year: 2026: 2.
2 papers.