# Decoding-time Realignment of Language Models

arXiv: https://arxiv.org/abs/2402.02992 · HTML: https://mlsanity.com/paper/2402.02992/

## Key facts

- Cited by 67 papers in the map (data through 14 Oct 2026).
- Placed in LLM preference optimization methods, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 40 of its 46 references that are in the map.
- First posted on arXiv 5 Feb 2024.

## Placed in

- LLM alignment / RLHF / preference optimisation / [LLM preference optimization methods](https://mlsanity.com/alignment/llm-preference-optimization-methods/)

## Builds on (most cited first, 30 of 40)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [Mistral 7B](https://mlsanity.com/paper/2310.06825/) (2023)
- [Learning to summarize from human feedback](https://mlsanity.com/paper/2009.01325/) (2020)
- [Fine-Tuning Language Models from Human Preferences](https://mlsanity.com/paper/1909.08593/) (2019)
- [High-Dimensional Continuous Control Using Generalized Advantage Estimation](https://mlsanity.com/paper/1506.02438/) (2015)
- [Fast Inference from Transformers via Speculative Decoding](https://mlsanity.com/paper/2211.17192/) (2022)
- [Concrete Problems in AI Safety](https://mlsanity.com/paper/1606.06565/) (2016)
- [Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization](https://mlsanity.com/paper/1808.08745/) (2018)
- [PaLM 2 Technical Report](https://mlsanity.com/paper/2305.10403/) (2023)
- [Accelerating Large Language Model Decoding with Speculative Sampling](https://mlsanity.com/paper/2302.01318/) (2023)
- [A General Theoretical Paradigm to Understand Learning from Human Preferences](https://mlsanity.com/paper/2310.12036/) (2023)
- [Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2307.15217/) (2023)
- [Contrastive Decoding: Open-ended Text Generation as Optimization](https://mlsanity.com/paper/2210.15097/) (2022)
- [Zephyr: Direct Distillation of LM Alignment](https://mlsanity.com/paper/2310.16944/) (2023)
- [SLiC-HF: Sequence Likelihood Calibration with Human Feedback](https://mlsanity.com/paper/2305.10425/) (2023)
- [The Effects of Reward Misspecification: Mapping and Mitigating Misaligned Models](https://mlsanity.com/paper/2201.03544/) (2022)
- [Statistical Rejection Sampling Improves Preference Optimization](https://mlsanity.com/paper/2309.06657/) (2023)
- [Personalized Soups: Personalized Large Language Model Alignment via Post-hoc Parameter Merging](https://mlsanity.com/paper/2310.11564/) (2023)
- [Is Reinforcement Learning (Not) for Natural Language Processing: Benchmarks, Baselines, and Building Blocks for Natural Language Policy Optimization](https://mlsanity.com/paper/2210.01241/) (2022)
- [Rewarded soups: towards Pareto-optimal alignment by interpolating weights fine-tuned on diverse rewards](https://mlsanity.com/paper/2306.04488/) (2023)
- [Nash Learning from Human Feedback](https://mlsanity.com/paper/2312.00886/) (2023)

## Built on by (most cited first, 30 of 67)

- [From r to Q^*: Your Language Model is Secretly a Q-Function](https://mlsanity.com/paper/2404.12358/) (2024)
- [Is Best-of-N the Best of Them? Coverage, Scaling, and Optimality in Inference-Time Alignment](https://mlsanity.com/paper/2503.21878/) (2025)
- [Decoding-Time Language Model Alignment with Multiple Objectives](https://mlsanity.com/paper/2406.18853/) (2024)
- [Harnessing Multiple Large Language Models: A Survey on LLM Ensemble](https://mlsanity.com/paper/2502.18036/) (2025)
- [SEAL: Safety-enhanced Aligned LLM Fine-tuning via Bilevel Data Selection](https://mlsanity.com/paper/2410.07471/) (2024)
- [PAD: Personalized Alignment of LLMs at Decoding-Time](https://mlsanity.com/paper/2410.04070/) (2024)
- [Transfer Q Star: Principled Decoding for LLM Alignment](https://mlsanity.com/paper/2405.20495/) (2024)
- [WARP: On the Benefits of Weight Averaged Rewarded Policies](https://mlsanity.com/paper/2406.16768/) (2024)
- [Conditional Language Policy: A General Framework for Steerable Multi-Objective Finetuning](https://mlsanity.com/paper/2407.15762/) (2024)
- [Cascade Reward Sampling for Efficient Decoding-Time Alignment](https://mlsanity.com/paper/2406.16306/) (2024)
- [Model Extrapolation Expedites Alignment](https://mlsanity.com/paper/2404.16792/) (2024)
- [Personality Alignment of Large Language Models](https://mlsanity.com/paper/2408.11779/) (2024)
- [Erasing Conceptual Knowledge from Language Models](https://mlsanity.com/paper/2410.02760/) (2024)
- [Amulet: ReAlignment During Test Time for Personalized Preference Adaptation of LLMs](https://mlsanity.com/paper/2502.19148/) (2025)
- [Spread Preference Annotation: Direct Preference Judgment for Efficient LLM Alignment](https://mlsanity.com/paper/2406.04412/) (2024)
- [The Crucial Role of Samplers in Online Direct Preference Optimization](https://mlsanity.com/paper/2409.19605/) (2024)
- [A Survey on Personalized and Pluralistic Preference Alignment in Large Language Models](https://mlsanity.com/paper/2504.07070/) (2025)
- [SimPER: A Minimalist Approach to Preference Alignment without Hyperparameters](https://mlsanity.com/paper/2502.00883/) (2025)
- [How to Merge Your Multimodal Models Over Time?](https://mlsanity.com/paper/2412.06712/) (2024)
- [Weak-to-Strong Preference Optimization: Stealing Reward from Weak Aligned Model](https://mlsanity.com/paper/2410.18640/) (2024)
- [Robust Multi-Objective Preference Alignment with Online DPO](https://mlsanity.com/paper/2503.00295/) (2025)
- [PILAF: Optimal Human Preference Sampling for Reward Modeling](https://mlsanity.com/paper/2502.04270/) (2025)
- [Towards a Unified View of Preference Learning for Large Language Models: A Survey](https://mlsanity.com/paper/2409.02795/) (2024)
- [Representation-Based Exploration for Language Models: From Test-Time to Post-Training](https://mlsanity.com/paper/2510.11686/) (2025)
- [On the Direction of RLVR Updates for LLM Reasoning: Identification and Exploitation](https://mlsanity.com/paper/2603.22117/) (2026)
- [Fast Large Language Model Collaborative Decoding via Speculation](https://mlsanity.com/paper/2502.01662/) (2025)
- [Unlocking Decoding-time Controllability: Gradient-Free Multi-Objective Alignment with Contrastive Prompts](https://mlsanity.com/paper/2408.05094/) (2024)
- [Robust Multi-Objective Controlled Decoding of Large Language Models](https://mlsanity.com/paper/2503.08796/) (2025)
- [ProxyThinker: Test-Time Guidance through Small Visual Reasoners](https://mlsanity.com/paper/2505.24872/) (2025)
- [Can DPO Learn Diverse Human Values? A Theoretical Scaling Law](https://mlsanity.com/paper/2408.03459/) (2024)

## Abstract (arXiv)

Aligning language models with human preferences is crucial for reducing errors and biases in these models. Alignment techniques, such as reinforcement learning from human feedback (RLHF), are typically cast as optimizing a tradeoff between human preference rewards and a proximity regularization term that encourages staying close to the unaligned model. Selecting an appropriate level of regularization is critical: insufficient regularization can lead to reduced model capabilities due to reward hacking, whereas excessive regularization hinders alignment. Traditional methods for finding the optimal regularization level require retraining multiple models with varying regularization strengths. This process, however, is resource-intensive, especially for large models. To address this challenge, we propose decoding-time realignment (DeRa), a simple method to explore and evaluate different regularization strengths in aligned models without retraining. DeRa enables control over the degree of alignment, allowing users to smoothly transition between unaligned and aligned models. It also enhances the efficiency of hyperparameter tuning by enabling the identification of effective regularization strengths using a validation dataset.