# Test-Time Alignment via Hypothesis Reweighting

arXiv: https://arxiv.org/abs/2412.08812 · HTML: https://mlsanity.com/paper/2412.08812/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Placed in Human values and AI feedback alignment, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 68 of its 88 references that are in the map.
- First posted on arXiv 11 Dec 2024.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Human values and AI feedback alignment](https://mlsanity.com/alignment/human-values-and-ai-feedback-alignment/)

## Builds on (most cited first, 30 of 68)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks](https://mlsanity.com/paper/1703.03400/) (2017)
- [Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning](https://mlsanity.com/paper/1506.02142/) (2015)
- [Parameter-Efficient Transfer Learning for NLP](https://mlsanity.com/paper/1902.00751/) (2019)
- [Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles](https://mlsanity.com/paper/1612.01474/) (2016)
- [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)
- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context](https://mlsanity.com/paper/2403.05530/) (2024)
- [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://mlsanity.com/paper/2101.03961/) (2021)
- [Prototypical Networks for Few-shot Learning](https://mlsanity.com/paper/1703.05175/) (2017)
- [Learning to summarize from human feedback](https://mlsanity.com/paper/2009.01325/) (2020)
- [Matching Networks for One Shot Learning](https://mlsanity.com/paper/1606.04080/) (2016)
- [Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters](https://mlsanity.com/paper/2408.03314/) (2024)
- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [Invariant Risk Minimization](https://mlsanity.com/paper/1907.02893/) (2019)
- [Mixtral of Experts](https://mlsanity.com/paper/2401.04088/) (2024)
- [GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding](https://mlsanity.com/paper/2006.16668/) (2020)
- [WILDS: A Benchmark of in-the-Wild Distribution Shifts](https://mlsanity.com/paper/2012.07421/) (2020)
- [Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift](https://mlsanity.com/paper/1906.02530/) (2019)
- [Making Pre-trained Language Models Better Few-shot Learners](https://mlsanity.com/paper/2012.15723/) (2020)
- [Distributionally Robust Neural Networks for Group Shifts: On the Importance of Regularization for Worst-Case Generalization](https://mlsanity.com/paper/1911.08731/) (2019)
- [Scaling Laws for Reward Model Overoptimization](https://mlsanity.com/paper/2210.10760/) (2022)
- [Exploration by Random Network Distillation](https://mlsanity.com/paper/1810.12894/) (2018)
- [Large Language Monkeys: Scaling Inference Compute with Repeated Sampling](https://mlsanity.com/paper/2407.21787/) (2024)
- [Gemma: Open Models Based on Gemini Research and Technology](https://mlsanity.com/paper/2403.08295/) (2024)
- [DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines](https://mlsanity.com/paper/2310.03714/) (2023)

## Built on by (most cited first, 11 of 11)

- [LoRe: Personalizing LLMs via Low-Rank Reward Modeling](https://mlsanity.com/paper/2504.14439/) (2025)
- [FSPO: Few-Shot Optimization of Synthetic Preferences Personalizes to Real Users](https://mlsanity.com/paper/2502.19312/) (2025)
- [MiCRo: Mixture Modeling and Context-aware Routing for Personalized Preference Learning](https://mlsanity.com/paper/2505.24846/) (2025)
- [Rethinking Diverse Human Preference Learning through Principal Component Analysis](https://mlsanity.com/paper/2502.13131/) (2025)
- [AdaBoN: Adaptive Best-of-N Alignment](https://mlsanity.com/paper/2505.12050/) (2025)
- [Capturing Individual Human Preferences with Reward Features](https://mlsanity.com/paper/2503.17338/) (2025)
- [PrefDisco: Benchmarking Proactive Personalized Reasoning](https://mlsanity.com/paper/2510.00177/) (2025)
- [POPI: Personalizing LLMs via Optimized Natural Language Preference Inference](https://mlsanity.com/paper/2510.17881/) (2025)
- [Sparse Mixture-of-Experts Reward Models Learn Interpretable and Specialized Experts for Personalized Preference Modeling](https://mlsanity.com/paper/2606.04284/) (2026)
- [Differentially Private Preference Data Synthesis for Large Language Model Alignment](https://mlsanity.com/paper/2605.30808/) (2026)
- [Beyond Routing Weights: Faithful Response-Level Interpretation of Mixture-of-Experts Reward Models via Contribution Contrast](https://mlsanity.com/paper/2608.06400/) (2026)

## Abstract (arXiv)

Reward models trained on aggregate preferences often fail to capture individual users' values, but existing adaptation methods such as fine-tuning or long-context conditioning are too costly for real-time personalization. We propose Hypothesis Reweighting (HyRe), which enables real-time personalization by reweighting ensemble members using just 1-5 labeled examples from the target user or domain. Our method builds on the empirical observation that when different heads capture different valid interpretations of preference data, reweighting them can substantially outperform uniform averaging. HyRe trains a single network with multiple prediction heads that capture different valid interpretations of preference data, then uses a Bayesian update to upweight the heads that best match the target user's preferences. This requires only a single forward pass with negligible (<1%) computational overhead, making it practical for inference-time personalization. We evaluate HyRe across diverse target preference distributions. With as few as five preference pairs per target distribution, HyRe surpasses state-of-the-art reward models on RewardBench at 2B and 8B scale and improves reward model accuracy by 20% across 32 personalization tasks.