Tag: Efficiency

All the articles with the tag "Efficiency".

• Does quantization affect models' performance on long-context tasks?

  Published: 2 Jun, 2025 |  at  11:34 AM

  

  87.84 🤔

  本文系统评估了量化对大型语言模型在长上下文任务中的性能影响，发现8-bit量化基本保持准确率（下降约0.8%），而4-bit量化导致显著损失（最高达59%），且影响因模型、任务和语言而异，强调了在长上下文和多语言场景下谨慎应用量化的必要性。

• EMORL: Ensemble Multi-Objective Reinforcement Learning for Efficient and Flexible LLM Fine-Tuning

  Published: 7 May, 2025 |  at  09:32 AM

  

  87.79 🤔

  本文提出EMORL框架，通过集成学习分别训练单目标模型并在隐藏状态层聚合，结合分层网格搜索优化权重，在咨询反思生成任务中实现了与传统方法相当的性能，同时显著提升了训练效率、可扩展性和解释性。

• Thought calibration: Efficient and confident test-time scaling

  Published: 28 May, 2025 |  at  11:22 AM

  

  87.79 🤔

  本文提出‘思想校准’方法，通过推理树抽象和轻量级探针动态决定语言模型推理终止时机，在分布内数据上减少高达60%的思考token，同时保持性能，并在分布外数据上实现20%的减少。

• Route to Reason: Adaptive Routing for LLM and Reasoning Strategy Selection

  Published: 28 May, 2025 |  at  11:21 AM

  

  87.78 🤔

  本文提出Route-To-Reason（RTR）框架，通过动态路由机制联合选择最优模型和推理策略，在多个推理任务上实现了更高的准确率和超过60%的token使用量减少，显著优化了性能与成本的权衡。

• Sparse-Group Boosting with Balanced Selection Frequencies: A Simulation-Based Approach and R Implementation

  Published: 8 May, 2025 |  at  10:25 AM

  

  87.75 🤔

  This paper introduces sparse-group boosting and a simulation-based group balancing algorithm within the 'sgboost' R package to mitigate variable selection bias in high-dimensional grouped data, demonstrating improved fairness and interpretability through simulations and ecological data analysis.