论文速读记录 | 2025.10

论文速读记录 | 2025.10


目录
  • Horizon Generalization in Reinforcement Learning
  • HIQL: Offline Goal-Conditioned RL with Latent States as Actions
  • Contrastive Preference Learning: Learning from Human Feedback without RL
  • Controlled Diversity with Preference: Towards Learning a Diverse Set of Desired Skills
  • Human-Aligned Skill Discovery Balancing Behaviour Exploration and Alignment
  • Few is More: Task-Efficient Skill-Discovery for Multi-Task Offline Multi-Agent Reinforcement Learning
  • SMAC-R1: The Emergence of Intelligence in Decision-Making Tasks
  • Efficient Off-Policy Meta-Reinforcement Learning via Probabilistic Context Variables
  • VariBAD: A Very Good Method for Bayes-Adaptive Deep RL via Meta-Learning
  • Rethinking Reward Modeling in Preference-based Large Language Model Alignment
  • DOPL: Direct Online Preference Learning for Restless Bandits with Preference Feedback
  • Fewer May Be Better: Enhancing Offline Reinforcement Learning with Reduced Dataset
  • Data Center Cooling System Optimization Using Offline Reinforcement Learning
  • SpikeLLM: Scaling up Spiking Neural Network to Large Language Models via Saliency-based Spiking
  • Rethinking Inverse Reinforcement Learning: from Data Alignment to Task Alignment
  • Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?
  • Why Distillation can Outperform Zero-RL: The Role of Flexible Reasoning
  • Thinkless: LLM Learns When to Think
  • Learning to Reason without External Rewards


Horizon Generalization in Reinforcement Learning

  • arxiv:https://arxiv.org/abs/2501.02709
  • website:https://horizon-generalization.github.io/
  • 来源:Benjamin Eysenbach 的新作,是一篇 arxiv paper,同学说有趣。
  • 主要内容:

HIQL: Offline Goal-Conditioned RL with Latent States as Actions

  • arxiv:https://arxiv.org/abs/2307.11949
  • website:https://seohong.me/projects/hiql/
  • 来源:合作者推荐的文章,好像也是 Benjamin Eysenbach 发表的。

Contrastive Preference Learning: Learning from Human Feedback without RL

  • arxiv:https://arxiv.org/abs/2310.13639
  • GitHub:https://github.com/jhejna/cpl
  • 来源:无意中搜到的文章,ICLR 2024,好像之前读过。
  • 主要内容:

Controlled Diversity with Preference: Towards Learning a Diverse Set of Desired Skills

  • arxiv:https://arxiv.org/abs/2303.04592
  • 来源:[mask]

Human-Aligned Skill Discovery Balancing Behaviour Exploration and Alignment

  • arxiv:https://arxiv.org/abs/2501.17431
  • 来源:[mask]

Few is More: Task-Efficient Skill-Discovery for Multi-Task Offline Multi-Agent Reinforcement Learning

  • arxiv:https://arxiv.org/abs/2502.08985
  • 来源:同学的最新工作。
  • 主要内容:
    • 这篇文章关注的 setting 是 offline multi-task MARL;特别的,agent 只在(比如说)三个人合作的场景上训练,然后就可以泛化到任意多个人合作的场景。同学讲的故事是,用 transformer 作为一个翻译器,把三个人的合作动作翻译为多个人的,感觉这个故事听起来非常好。

SMAC-R1: The Emergence of Intelligence in Decision-Making Tasks

  • arxiv:https://arxiv.org/abs/2410.16024
  • 来源:在知乎看到的,但现在知乎帖子好像找不到了)
  • 主要内容:
    • 用 LLM 生成打 smac 的 python 决策树代码。
    • 具体 method:

Efficient Off-Policy Meta-Reinforcement Learning via Probabilistic Context Variables

  • arxiv:https://arxiv.org/abs/1903.08254
  • 来源:[mask]
  • 主要内容:
    • 这篇文章提出了 PERAL 方法。

VariBAD: A Very Good Method for Bayes-Adaptive Deep RL via Meta-Learning

  • arxiv:https://arxiv.org/abs/1910.08348
  • 来源:[mask]
  • 主要内容:
    • 这篇文章提出了 VariBAD 方法。

Rethinking Reward Modeling in Preference-based Large Language Model Alignment

  • arxiv:https://arxiv.org/abs/2411.04991
  • OpenReview:https://openreview.net/forum?id=rfdblE10qm
  • 来源:ICLR 2025 oral。
  • 主要内容:
    • 这篇文章关注 LLM 的 RLHF。据说不采用 bradley-terry model 来建模 reward model,而是直接训一个分类器,学习一个 (x,y) 是好的还剩坏的,然后使用分类器的概率 logit 作为 RLHF 的 reward。
    • 是否使用了非成对的比较 \((x_1, y_1^+, x_2, y_2^-)\),而非把成对比较 \((x, y^+, y^-)\) 打乱(?)
    • 实验是否过于 toy(?)理论大概说了什么(?)

DOPL: Direct Online Preference Learning for Restless Bandits with Preference Feedback

  • arxiv:https://arxiv.org/abs/2410.05527
  • open review:https://openreview.net/forum?id=2iYVBqRHK4
  • 来源:合作者推荐的文章。
  • 主要内容:
    • preference-based index policy(?)

Fewer May Be Better: Enhancing Offline Reinforcement Learning with Reduced Dataset

  • 来源:师兄的文章。

Data Center Cooling System Optimization Using Offline Reinforcement Learning

  • arxiv:https://arxiv.org/pdf/2501.15085
  • 来源:xianyuan zhan 组的新文章。
  • 主要内容:
    • T-symmetry。

SpikeLLM: Scaling up Spiking Neural Network to Large Language Models via Saliency-based Spiking

  • arxiv:https://arxiv.org/abs/2407.04752
  • 来源:师兄推荐的神秘文章,ICLR 2025 poster。

Rethinking Inverse Reinforcement Learning: from Data Alignment to Task Alignment

  • arxiv:https://arxiv.org/abs/2410.23680
  • 来源:偶然看到的文章。

Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?

  • 来源:师兄偶然提到,系里其他人的文章。

Why Distillation can Outperform Zero-RL: The Role of Flexible Reasoning

  • arxiv:https://arxiv.org/abs/2505.21067
  • 来源:偶然看到的文章。

Thinkless: LLM Learns When to Think

  • arxiv:https://arxiv.org/abs/2505.13379
  • 来源:偶然看到的文章。

Learning to Reason without External Rewards

  • arxiv:https://arxiv.org/abs/2505.19590
  • 来源:偶然看到的文章。