# MACPO: Weak-to-Strong Alignment via Multi-Agent Contrastive Preference Optimization

arXiv: https://arxiv.org/abs/2410.07672 · HTML: https://mlsanity.com/paper/2410.07672/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- Placed in Reinforcement learning for LLM agents, one of 16 directions in LLM agents and tool use.
- Builds on 82 of its 96 references that are in the map.
- First posted on arXiv 10 Oct 2024.

## Placed in

- LLM agents and tool use / [Reinforcement learning for LLM agents](https://mlsanity.com/agents/reinforcement-learning-for-llm-agents/)
- LLM alignment / RLHF / preference optimisation / [LLM preference optimization methods](https://mlsanity.com/alignment/llm-preference-optimization-methods/)

## Builds on (most cited first, 30 of 82)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [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)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [Generative Agents: Interactive Simulacra of Human Behavior](https://mlsanity.com/paper/2304.03442/) (2023)
- [Universal and Transferable Adversarial Attacks on Aligned Language Models](https://mlsanity.com/paper/2307.15043/) (2023)
- [Constitutional AI: Harmlessness from AI Feedback](https://mlsanity.com/paper/2212.08073/) (2022)
- [MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework](https://mlsanity.com/paper/2308.00352/) (2023)
- [Improving Factuality and Reasoning in Language Models through Multiagent Debate](https://mlsanity.com/paper/2305.14325/) (2023)
- [LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models](https://mlsanity.com/paper/2403.13372/) (2024)
- [Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generation](https://mlsanity.com/paper/2305.01210/) (2023)
- [Large Language Models are not Fair Evaluators](https://mlsanity.com/paper/2305.17926/) (2023)
- [A General Language Assistant as a Laboratory for Alignment](https://mlsanity.com/paper/2112.00861/) (2021)
- [SimPO: Simple Preference Optimization with a Reference-Free Reward](https://mlsanity.com/paper/2405.14734/) (2024)
- [ChatDev: Communicative Agents for Software Development](https://mlsanity.com/paper/2307.07924/) (2023)
- [Large Language Model based Multi-Agents: A Survey of Progress and Challenges](https://mlsanity.com/paper/2402.01680/) (2024)
- [AlpacaFarm: A Simulation Framework for Methods that Learn from Human Feedback](https://mlsanity.com/paper/2305.14387/) (2023)
- [WizardCoder: Empowering Code Large Language Models with Evol-Instruct](https://mlsanity.com/paper/2306.08568/) (2023)
- [Self-Rewarding Language Models](https://mlsanity.com/paper/2401.10020/) (2024)
- [Safe RLHF: Safe Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2310.12773/) (2023)
- [WizardMath: Empowering Mathematical Reasoning for Large Language Models via Reinforced Evol-Instruct](https://mlsanity.com/paper/2308.09583/) (2023)
- [AgentVerse: Facilitating Multi-Agent Collaboration and Exploring Emergent Behaviors](https://mlsanity.com/paper/2308.10848/) (2023)
- [Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Models](https://mlsanity.com/paper/2401.01335/) (2024)
- [CCNet: Extracting High Quality Monolingual Datasets from Web Crawl Data](https://mlsanity.com/paper/1911.00359/) (2019)
- [Reinforced Self-Training (ReST) for Language Modeling](https://mlsanity.com/paper/2308.08998/) (2023)
- [Is ChatGPT a General-Purpose Natural Language Processing Task Solver?](https://mlsanity.com/paper/2302.06476/) (2023)

## Built on by (most cited first, 22 of 22)

- [EvoScientist: Towards Multi-Agent Evolving AI Scientists for End-to-End Scientific Discovery](https://mlsanity.com/paper/2603.08127/) (2026)
- [DeepShop: A Benchmark for Deep Research Shopping Agents](https://mlsanity.com/paper/2506.02839/) (2025)
- [Heterogeneous Swarms: Jointly Optimizing Model Roles and Weights for Multi-LLM Systems](https://mlsanity.com/paper/2502.04510/) (2025)
- [Uni-X: Mitigating Modality Conflict with a Two-End-Separated Architecture for Unified Multimodal Models](https://mlsanity.com/paper/2509.24365/) (2025)
- [The Road to Artificial SuperIntelligence: A Comprehensive Survey of Superalignment](https://mlsanity.com/paper/2412.16468/) (2024)
- [Co-Evolving Agents: Learning from Failures as Hard Negatives](https://mlsanity.com/paper/2511.22254/) (2025)
- [Well Begun is Half Done: Low-resource Preference Alignment by Weak-to-Strong Decoding](https://mlsanity.com/paper/2506.07434/) (2025)
- [EvoIdeator: Evolving Scientific Ideas through Checklist-Grounded Reinforcement Learning](https://mlsanity.com/paper/2603.21728/) (2026)
- [Weak-to-Strong Generalization via Bregman Bias-Variance Decomposition](https://mlsanity.com/paper/2505.24313/) (2025)
- [How to Mitigate Overfitting in Weak-to-strong Generalization?](https://mlsanity.com/paper/2503.04249/) (2025)
- [Weak-to-Strong Generalization with Failure Trajectories: A Tree-based Approach to Elicit Optimal Policy in Strong Models](https://mlsanity.com/paper/2507.18858/) (2025)
- [Super Co-alignment of Human and AI for Sustainable Symbiotic Society](https://mlsanity.com/paper/2504.17404/) (2025)
- [Contrastive Weak-to-strong Generalization](https://mlsanity.com/paper/2510.07884/) (2025)
- [Research Superalignment Should Advance Now with Alternating Competence and Conformity Optimization](https://mlsanity.com/paper/2503.07660/) (2025)
- [Bridging the Capability Gap: Joint Alignment Tuning for Harmonizing LLM-based Multi-Agent Systems](https://mlsanity.com/paper/2509.09629/) (2025)
- [On the Blessing of Pre-training in Weak-to-Strong Generalization](https://mlsanity.com/paper/2605.05710/) (2026)
- [SearchJev: A Fast and Calibrated System-1 Model for Search Agents](https://mlsanity.com/paper/2610.05107/) (2026)
- [ATLAS: A Multi-LLM Training Framework for EvoDPO with Adaptive Reference Evolution](https://mlsanity.com/paper/2602.02709/) (2026)
- [Evaluating Risks in Weak-to-Strong Alignment: A Bias-Variance Perspective](https://mlsanity.com/paper/2604.25077/) (2026)
- [When In-Distribution Gains Fail: Evaluating Weak-to-Strong Reward Models under Preference Shift](https://mlsanity.com/paper/2605.25629/) (2026)
- [StepGuard: Guarding Web Navigation via Single-Step Calibration](https://mlsanity.com/paper/2606.17871/) (2026)
- [SwarmRSI: Recursive Self-Improvement through Evolving Scientific Agent Swarm](https://mlsanity.com/paper/2610.20431/) (2026)

## Abstract (arXiv)

As large language models (LLMs) are rapidly advancing and achieving near-human capabilities on specific tasks, aligning them with human values is becoming more urgent. In scenarios where LLMs outperform humans, we face a weak-to-strong alignment problem where we need to effectively align strong student LLMs through weak supervision generated by weak teachers. Existing alignment methods mainly focus on strong-to-weak alignment and self-alignment settings, and it is impractical to adapt them to the much harder weak-to-strong alignment setting. To fill this gap, we propose a multi-agent contrastive preference optimization (MACPO) framework. MACPO facilitates weak teachers and strong students to learn from each other by iteratively reinforcing unfamiliar positive behaviors while penalizing familiar negative ones. To get this, we devise a mutual positive behavior augmentation strategy to encourage weak teachers and strong students to learn from each other's positive behavior and further provide higher quality positive behavior for the next iteration. Additionally, we propose a hard negative behavior construction strategy to induce weak teachers and strong students to generate familiar negative behavior by fine-tuning on negative behavioral data. Experimental results on the HH-RLHF and PKU-SafeRLHF datasets, evaluated using both automatic metrics and human judgments, demonstrate that MACPO simultaneously improves the alignment performance of strong students and weak teachers. Moreover, as the number of weak teachers increases, MACPO achieves better weak-to-strong alignment performance through more iteration optimization rounds.