# Preference Goal Tuning: Post-Training as Latent Control for Frozen Policies

arXiv: https://arxiv.org/abs/2412.02125 · HTML: https://mlsanity.com/paper/2412.02125/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Reward modeling and RLHF, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 33 of its 50 references that are in the map.
- First posted on arXiv 3 Dec 2024.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Reward modeling and RLHF](https://mlsanity.com/alignment/reward-modeling-and-rlhf/)

## Builds on (most cited first, 30 of 33)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [The Power of Scale for Parameter-Efficient Prompt Tuning](https://mlsanity.com/paper/2104.08691/) (2021)
- [OpenVLA: An Open-Source Vision-Language-Action Model](https://mlsanity.com/paper/2406.09246/) (2024)
- [π0: A Vision-Language-Action Flow Model for General Robot Control](https://mlsanity.com/paper/2410.24164/) (2024)
- [Fine-Tuning Language Models from Human Preferences](https://mlsanity.com/paper/1909.08593/) (2019)
- [Gradient Episodic Memory for Continual Learning](https://mlsanity.com/paper/1706.08840/) (2017)
- [Octo: An Open-Source Generalist Robot Policy](https://mlsanity.com/paper/2405.12213/) (2024)
- [LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learning](https://mlsanity.com/paper/2306.03310/) (2023)
- [BitFit: Simple Parameter-efficient Fine-tuning for Transformer-based Masked Language-models](https://mlsanity.com/paper/2106.10199/) (2021)
- [SimPO: Simple Preference Optimization with a Reference-Free Reward](https://mlsanity.com/paper/2405.14734/) (2024)
- [A General Theoretical Paradigm to Understand Learning from Human Preferences](https://mlsanity.com/paper/2310.12036/) (2023)
- [RDT-1B: a Diffusion Foundation Model for Bimanual Manipulation](https://mlsanity.com/paper/2410.07864/) (2024)
- [BC-Z: Zero-Shot Task Generalization with Robotic Imitation Learning](https://mlsanity.com/paper/2202.02005/) (2022)
- [MineDojo: Building Open-Ended Embodied Agents with Internet-Scale Knowledge](https://mlsanity.com/paper/2206.08853/) (2022)
- [SLiC-HF: Sequence Likelihood Calibration with Human Feedback](https://mlsanity.com/paper/2305.10425/) (2023)
- [Describe, Explain, Plan and Select: Interactive Planning with Large Language Models Enables Open-World Multi-Task Agents](https://mlsanity.com/paper/2302.01560/) (2023)
- [Video PreTraining (VPT): Learning to Act by Watching Unlabeled Online Videos](https://mlsanity.com/paper/2206.11795/) (2022)
- [Prismatic VLMs: Investigating the Design Space of Visually-Conditioned Language Models](https://mlsanity.com/paper/2402.07865/) (2024)
- [MimicPlay: Long-Horizon Imitation Learning by Watching Human Play](https://mlsanity.com/paper/2302.12422/) (2023)
- [MineRL: A Large-Scale Dataset of Minecraft Demonstrations](https://mlsanity.com/paper/1907.13440/) (2019)
- [JARVIS-1: Open-World Multi-task Agents with Memory-Augmented Multimodal Language Models](https://mlsanity.com/paper/2311.05997/) (2023)
- [Calibrating Sequence likelihood Improves Conditional Language Generation](https://mlsanity.com/paper/2210.00045/) (2022)
- [GRAPE: Generalizing Robot Policy via Preference Alignment](https://mlsanity.com/paper/2411.19309/) (2024)
- [GROOT: Learning to Follow Instructions by Watching Gameplay Videos](https://mlsanity.com/paper/2310.08235/) (2023)
- [Open-World Multi-Task Control Through Goal-Aware Representation Learning and Adaptive Horizon Prediction](https://mlsanity.com/paper/2301.10034/) (2023)
- [End-to-End Navigation with Vision Language Models: Transforming Spatial Reasoning into Question-Answering](https://mlsanity.com/paper/2411.05755/) (2024)

## Built on by (most cited first, 3 of 3)

- [JARVIS-VLA: Post-Training Large-Scale Vision Language Models to Play Visual Games with Keyboards and Mouse](https://mlsanity.com/paper/2503.16365/) (2025)
- [Open-World Skill Discovery from Unsegmented Demonstrations](https://mlsanity.com/paper/2503.10684/) (2025)
- [Self-Guided Action Diffusion](https://mlsanity.com/paper/2508.12189/) (2025)

## Abstract (arXiv)

Goal-conditioned policies enable decision-making models to execute diverse behaviors based on specified goals, yet their downstream performance is often highly sensitive to the choice of instructions or prompts. To bypass the limitations of discrete text prompts, we formulate post-training adaptation as a latent control problem, where the goal embedding serves as a continuous control variable to modulate the behavior of a frozen policy. We propose Preference Goal Tuning (PGT), a framework that optimizes this latent control variable to align the induced trajectory distribution with task preferences. Unlike standard fine-tuning that updates policy parameters, PGT keeps the policy frozen and updates only the latent goal using a trajectory-level preference objective. This approach essentially searches for the optimal conditioning input that maximizes the likelihood of preferred behaviors while suppressing undesirable ones. We evaluate PGT on the Minecraft SkillForge benchmark across 17 tasks. With minimal data, PGT achieves average relative improvements of 72.0% and 81.6% on two foundation policies, consistently outperforming expert-crafted prompts. Crucially, by decoupling task alignment (latent goal) from physical dynamics (frozen policy), PGT surpasses full fine-tuning by 13.4% in out-of-distribution settings, demonstrating superior robustness and generalization.