cs.CL（2025-03-28）

📊 共 21 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (20 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (20 篇)

#题目一句话要点标签🔗⭐
1 Exploiting Mixture-of-Experts Redundancy Unlocks Multimodal Generative Abilities 利用MoE冗余容量，实现参数高效的多模态生成能力扩展 large language model multimodal
2 Historical Ink: Exploring Large Language Models for Irony Detection in 19th-Century Spanish 利用大型语言模型提升19世纪西班牙语报刊讽刺检测 large language model
3 Firm or Fickle? Evaluating Large Language Models Consistency in Sequential Interactions 提出PWC指标、MT-Consistency基准和CARG框架，提升LLM在多轮交互中的一致性 large language model ✅
4 Penrose Tiled Low-Rank Compression and Section-Wise Q&A Fine-Tuning: A General Framework for Domain-Specific Large Language Model Adaptation 提出Penrose平铺低秩压缩与分段问答微调框架，用于领域特定大语言模型高效适配。 large language model
5 Generalization Bias in Large Language Model Summarization of Scientific Research 揭示大型语言模型在科学研究总结中存在的过度泛化偏差 large language model
6 Negation: A Pink Elephant in the Large Language Models' Room? 提出NoFEVER-ML和NoSNLI-ML数据集，评估并提升LLM在否定推理上的能力 large language model
7 Token-Driven GammaTune: Adaptive Calibration for Enhanced Speculative Decoding 提出GammaTune以解决大语言模型推理效率问题 large language model
8 Bridging the Dimensional Chasm: Uncover Layer-wise Dimensional Reduction in Transformers through Token Correlation 通过Token相关性揭示Transformer层维度缩减现象，弥合高维计算与低维语义鸿沟 large language model
9 Understanding Inequality of LLM Fact-Checking over Geographic Regions with Agent and Retrieval models 揭示LLM事实核查在不同地理区域上的不平等性，并分析Agent和检索模型的影响 large language model
10 Evaluating LLM-based Agents for Multi-Turn Conversations: A Survey 综述：评估基于LLM的Agent在多轮对话中的性能 large language model
11 FRASE: Structured Representations for Generalizable SPARQL Query Generation FRASE利用框架语义角色标注提升SPARQL查询生成泛化能力 large language model
12 Beyond Single-Sentence Prompts: Upgrading Value Alignment Benchmarks with Dialogues and Stories 提出对话与故事型基准，提升LLM价值观对齐评估的鲁棒性 large language model
13 Resona: Improving Context Copying in Linear Recurrence Models with Retrieval Resona：通过检索增强线性循环模型中的上下文复制能力 large language model
14 Learning to Reason for Long-Form Story Generation 提出基于可验证奖励的强化学习方法，用于提升长文本故事生成的推理能力。 large language model
15 WorkTeam: Constructing Workflows from Natural Language with Multi-Agents 提出WorkTeam多智能体框架，解决企业级自然语言到工作流构建难题。 large language model
16 Supposedly Equivalent Facts That Aren't? Entity Frequency in Pre-training Induces Asymmetry in LLMs 揭示LLM中实体频率偏差导致逻辑等价事实识别的非对称性 large language model
17 SKDU at De-Factify 4.0: Natural Language Features for AI-Generated Text-Detection SKDU提出一种基于自然语言特征的AI生成文本检测流水线方法，并在De-Factify 4.0数据集上验证。 large language model
18 A Refined Analysis of Massive Activations in LLMs 针对LLM中大规模激活值，提出混合缓解策略以平衡性能与激活抑制 large language model ✅
19 MultiClaimNet: A Massively Multilingual Dataset of Fact-Checked Claim Clusters 提出MultiClaimNet，一个大规模多语种事实核查声明聚类数据集，促进高效事实核查。 large language model
20 EdgeInfinite: A Memory-Efficient Infinite-Context Transformer for Edge Devices EdgeInfinite：面向边缘设备的内存高效无限上下文Transformer large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (1 篇)

#题目一句话要点标签🔗⭐
21 A Review of Incorporating Psychological Theories in LLMs 综述心理学理论在大型语言模型中的应用，旨在促进类人认知与行为的建模。 reinforcement learning large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页