cs.CL（2025-03-14）

📊 共 24 篇论文 | 🔗 5 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (22 🔗5) 支柱二：RL算法与架构 (RL & Architecture) (2)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (22 篇)

#题目一句话要点标签🔗⭐
1 Don't Take Things Out of Context: Attention Intervention for Enhancing Chain-of-Thought Reasoning in Large Language Models 提出注意力干预方法FAI，增强大语言模型中的思维链推理能力 large language model chain-of-thought
2 Unlocking General Long Chain-of-Thought Reasoning Capabilities of Large Language Models via Representation Engineering 提出GLoRE：通过表征工程解锁大语言模型通用长链推理能力 large language model chain-of-thought
3 Exploring the Potential of Large Multimodal Models as Effective Alternatives for Pronunciation Assessment 探索大型多模态模型在发音评估中的应用潜力，以GPT-4o为例。 large language model multimodal
4 Evaluating the Process Modeling Abilities of Large Language Models -- Preliminary Foundations and Results 评估大语言模型的过程建模能力：初步基础与结果分析 large language model
5 Agent-Enhanced Large Language Models for Researching Political Institutions 提出Agentic RAG，增强LLM在政治机构研究中的数据处理与分析能力 large language model
6 High-Dimensional Interlingual Representations of Large Language Models 提出跨语言表征框架，评估多语言LLM中跨语言对齐并提升跨语言泛化能力。 large language model
7 TigerLLM - A Family of Bangla Large Language Models TigerLLM：构建并开源了一系列高性能孟加拉语大型语言模型 large language model
8 Relevance Isn't All You Need: Scaling RAG Systems With Inference-Time Compute Via Multi-Criteria Reranking REBEL：通过多标准重排序和推理时计算扩展RAG系统，提升检索相关性和答案质量。 large language model chain-of-thought ✅
9 RONA: Pragmatically Diverse Image Captioning with Coherence Relations 提出RONA，利用连贯关系提升多模态大语言模型生成图像描述的多样性。 large language model ✅
10 Resolving UnderEdit & OverEdit with Iterative & Neighbor-Assisted Model Editing 提出迭代与邻域辅助模型编辑方法，解决LLM知识更新中的欠编辑与过度编辑问题 large language model ✅
11 Annotating Scientific Uncertainty: A comprehensive model using linguistic patterns and comparison with existing approaches UnScientify：利用语言模式检测科学文本中的不确定性，性能优于大型语言模型。 large language model
12 AIstorian lets AI be a historian: A KG-powered multi-agent system for accurate biography generation AIstorian：基于知识图谱的多Agent系统，用于生成准确的历史人物传记 large language model ✅
13 RAG-KG-IL: A Multi-Agent Hybrid Framework for Reducing Hallucinations and Enhancing LLM Reasoning through RAG and Incremental Knowledge Graph Learning Integration RAG-KG-IL框架通过融合RAG与增量知识图谱学习，提升LLM推理能力并减少幻觉 large language model
14 LAG-MMLU: Benchmarking Frontier LLM Understanding in Latvian and Giriama LAG-MMLU：在拉脱维亚语和吉里亚玛语中评估前沿LLM的理解能力 large language model
15 OpeNLGauge: An Explainable Metric for NLG Evaluation with Open-Weights LLMs OpeNLGauge：一种基于开源LLM的、可解释的NLG评估指标。 large language model
16 Bridging the LLM Accessibility Divide? Performance, Fairness, and Cost of Closed versus Open LLMs for Automated Essay Scoring 对比闭源与开源LLM在自动作文评分中的表现、公平性和成本，揭示开源模型的潜力。 large language model
17 Key, Value, Compress: A Systematic Exploration of KV Cache Compression Techniques 系统性探索KV缓存压缩技术，提升长文本LLM推理效率 large language model
18 Modeling Subjectivity in Cognitive Appraisal with Language Models 利用语言模型建模认知评估中的主观性，提升人机交互理解。 large language model
19 Are formal and functional linguistic mechanisms dissociated in language models? 研究揭示：大型语言模型中形式语言和功能语言机制并非完全分离且缺乏统一的形式语言网络。 large language model
20 GNNs as Predictors of Agentic Workflow Performances 提出FLORA-Bench，利用GNN预测Agentic Workflow性能，优化LLM调用。 large language model ✅
21 Palette of Language Models: A Solver for Controlled Text Generation 提出一种基于概率论和互信息最小化的语言模型调色板，用于可控文本生成。 large language model
22 Towards Extreme Pruning of LLMs with Plug-and-Play Mixed Sparsity 提出一种即插即用的混合稀疏度剪枝方法，用于大语言模型的极端压缩。 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (2 篇)

#题目一句话要点标签🔗⭐
23 BriLLM: Brain-inspired Large Language Model BriLLM：提出脑启发大语言模型，通过信号全连接流动学习实现通用人工智能。 world model representation learning semantic mapping
24 Rule-Guided Feedback: Enhancing Reasoning by Enforcing Rule Adherence in Large Language Models 提出规则引导反馈（RGF）框架，通过规则约束提升大语言模型推理能力 teacher-student large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页