用元强化学习让大模型自动适配复杂约束，提速一倍还省电。
Adaptive Constraint Propagation: Scaling Structured Inference for Large Language Models via Meta-Reinforcement Learning
- 通过元强化学习训练通用约束传播策略，跨任务无需重训。
- 比现有加速方案快1.5到2倍，精度损失小于0.2%。
- 适合需要快速部署多语言结构化输出的场景，如数据抽取、逻辑推理。
大语言模型在生成过程中越来越依赖结构化推理，如JSON模式校验或跨语言解析，要求输出满足复杂约束。本文提出MetaJuLS，一种基于元强化学习的方法，可学习适用于多种语言和任务的通用约束传播策略，无需针对具体任务重新训练。将结构化推理建模为自适应约束传播，并使用图注意力网络结合元学习进行训练，使得在通用依赖（Universal Dependencies）跨10种语言及受约束生成任务（LogicBench、GSM8K-Constrained）中实现快速跨领域迁移：在英语解析上训练的策略仅需5–10步梯度更新（5–15秒），即可适应新语言与新任务，相比传统任务特训节省数小时。机制分析显示，该策略发现了类人类的解析策略（如先易后难）以及非直观的新启发式规则。通过减少大模型推理中的传播步骤，MetaJuLS直接降低推理碳足迹，助力绿色人工智能。
原文摘要 · Abstract (English)
Large language models increasingly require structured inference, from JSON schema enforcement to multi-lingual parsing, where outputs must satisfy complex constraints. We introduce MetaJuLS, a meta-reinforcement learning approach that learns universal constraint propagation policies applicable across languages and tasks without task-specific retraining. By formulating structured inference as adaptive constraint propagation and training a Graph Attention Network with meta-learning, MetaJuLS achieves 1.5--2.0$\times$ speedups over GPU-optimized baselines while maintaining within 0.2\% accuracy of state-of-the-art parsers. On Universal Dependencies across 10 languages and LLM-constrained generation (LogicBench, GSM8K-Constrained), MetaJuLS demonstrates rapid cross-domain adaptation: a policy trained on English parsing adapts to new languages and tasks with 5--10 gradient steps (5--15 seconds) rather than requiring hours of task-specific training. Mechanistic analysis reveals the policy discovers human-like parsing strategies (easy-first) and novel non-intuitive heuristics. By reducing propagation steps in LLM deployments, MetaJuLS contributes to Green AI by directly reducing inference carbon footprint.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。