cs.LG（2024-10-29）
📊 共 7 篇论文
🎯 兴趣领域导航
🔬 支柱九：具身大模型 (Embodied Foundation Models) (4 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 1 | SVIP: Towards Verifiable Inference of Open-source Large Language Models | SVIP：面向开源大语言模型的可验证推理方案，保障用户权益。 | large language model | ||
| 2 | Are Large-Language Models Graph Algorithmic Reasoners? | MAGMA基准测试揭示LLM在图算法推理上的不足，强调高级提示工程的必要性。 | large language model | ||
| 3 | BenchAgents: Multi-Agent Systems for Structured Benchmark Creation | BenchAgents：利用多智能体系统自动创建结构化评测基准 | large language model | ||
| 4 | Unlearning as multi-task optimization: A normalized gradient difference approach with an adaptive learning rate | 提出NGDiff算法，通过归一化梯度差异和自适应学习率优化LLM的不可学习性。 | large language model |
🔬 支柱二：RL算法与架构 (RL & Architecture) (3 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 5 | Fourier Head: Helping Large Language Models Learn Complex Probability Distributions | 提出傅里叶头，增强LLM建模复杂概率分布的能力，提升非语言token序列建模效果。 | decision transformer large language model foundation model | ||
| 6 | Solving Minimum-Cost Reach Avoid using Reinforcement Learning | 提出RC-PPO算法，解决最小成本可达-避障强化学习问题 | reinforcement learning PPO | ||
| 7 | A Large Recurrent Action Model: xLSTM enables Fast Inference for Robotics Tasks | 提出基于xLSTM的大型循环动作模型LRAM，加速机器人任务推理。 | reinforcement learning Mamba |