Tag: Reinforcement Learning

All the articles with the tag "Reinforcement Learning".

• StreamRL: Scalable, Heterogeneous, and Elastic RL for LLMs with Disaggregated Stream Generation

  Published: 4 May, 2025 |  at  04:29 PM

  

  79.93 🤔

  本文提出 StreamRL 框架，通过分离式流生成架构优化 RL 训练，解决了流水线和偏斜气泡问题，提高了 LLMs RL 训练的吞吐量和成本效率。

• Toward Efficient Exploration by Large Language Model Agents

  Published: 4 May, 2025 |  at  04:31 PM

  

  79.45 🤔

  本文通过使用 LLMs 显式实现后验采样 RL 算法，显著提高了 LLMs 代理在自然语言环境中的探索效率，同时保留了经典算法的统计性能优势。

• Reinforced MLLM: A Survey on RL-Based Reasoning in Multimodal Large Language Models

  Published: 7 May, 2025 |  at  08:42 AM

  

  78.41 🤔

  本文系统综述了基于强化学习的推理方法在多模态大语言模型（MLLMs）中的进展，分析了算法设计、奖励机制及应用，揭示了跨模态推理和奖励稀疏性等挑战，并提出了分层奖励和交互式RL等未来方向。

• Latent Preference Coding: Aligning Large Language Models via Discrete Latent Codes

  Published: 12 May, 2025 |  at  11:14 AM

  

  76.90 🤔

  This paper introduces Latent Preference Coding (LPC), a framework that uses discrete latent codes to model multifaceted human preferences, consistently improving the performance of offline alignment algorithms like DPO, SimPO, and IPO across multiple LLMs and benchmarks.

• Stop Overthinking: A Survey on Efficient Reasoning for Large Language Models

  Published: 4 May, 2025 |  at  04:26 PM

  

  76.52 🤔

  本文首次系统调查了大型语言模型高效推理的进展，通过分类模型、输出和提示-based方法，探讨了减少"过度思考"现象的策略，以优化计算效率并保持推理能力。