cs.CL（2025-03-30）

📊 共 21 篇论文

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (18) 支柱二：RL算法与架构 (RL & Architecture) (3)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (18 篇)

#题目一句话要点标签🔗⭐
1 Question-Aware Knowledge Graph Prompting for Enhancing Large Language Models 提出问题感知知识图谱提示方法，增强大语言模型在知识密集型问答任务中的表现 large language model
2 Extracting Patient History from Clinical Text: A Comparative Study of Clinical Large Language Models 评估临床大语言模型在病史实体抽取中的性能，并分析文本特征对模型准确率的影响 large language model
3 Multi-Stakeholder Disaster Insights from Social Media Using Large Language Models 利用大型语言模型从社交媒体提取多方利益相关者的灾害洞察 large language model
4 Beyond the Reported Cutoff: Where Large Language Models Fall Short on Financial Knowledge 评估大语言模型在金融知识上的局限性：超越报告截止日期后的表现 large language model
5 When LLM Therapists Become Salespeople: Evaluating Large Language Models for Ethical Motivational Interviewing 评估大型语言模型在伦理动机访谈中的应用，发现其伦理意识不足并提出改进方案 large language model
6 Evolutionary Prompt Optimization Discovers Emergent Multimodal Reasoning Strategies in Vision-Language Models 提出进化提示优化框架，提升视觉-语言模型的多模态推理能力。 multimodal
7 PromptDistill: Query-based Selective Token Retention in Intermediate Layers for Efficient Large Language Model Inference PromptDistill：一种基于查询的选择性token保留方法，用于高效的大语言模型推理。 large language model
8 CrossWordBench: Evaluating the Reasoning Capabilities of LLMs and LVLMs with Controllable Puzzle Generation CrossWordBench：提出可控填字游戏生成框架，评估LLM和LVLM的推理能力 large language model multimodal
9 Order Independence With Finetuning 通过微调提升LLM的顺序无关性，解决多项选择题中的位置偏见问题 large language model
10 Distill-C: Enhanced NL2SQL via Distilled Customization with LLMs 提出Distill-C框架，通过蒸馏定制提升LLM在NL2SQL任务中的性能与效率。 large language model
11 If an LLM Were a Character, Would It Know Its Own Story? Evaluating Lifelong Learning in LLMs 提出LIFESTATE-BENCH评估LLM的终身学习能力，揭示其在状态保持和记忆方面的挑战。 large language model
12 Mixture of Routers 提出混合路由（MoR）方法，提升LoRA微调大语言模型的性能和专家路由的均衡性。 large language model
13 RARE: Retrieval-Augmented Reasoning Modeling RARE：提出检索增强推理建模，解决LLM领域知识幻觉和推理能力不足问题 large language model
14 SCORE: Story Coherence and Retrieval Enhancement for AI Narratives 提出SCORE框架，增强AI生成叙事的连贯性和检索能力，解决故事一致性问题。 large language model
15 Discovering Knowledge Deficiencies of Language Models on Massive Knowledge Base 提出随机误差上升（SEA）框架，高效发现大规模知识库中语言模型的知识缺陷。 large language model
16 Not All LoRA Parameters Are Essential: Insights on Inference Necessity 提出LoRA层剪枝方法，通过识别关键层提升LLM推理效率与性能 large language model
17 Focus Directions Make Your Language Models Pay More Attention to Relevant Contexts 提出焦点方向，提升长文本语言模型对相关上下文的关注度 large language model
18 Cocktail: Chunk-Adaptive Mixed-Precision Quantization for Long-Context LLM Inference Cocktail：针对长文本LLM推理的块自适应混合精度量化 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🔗⭐
19 FeRG-LLM : Feature Engineering by Reason Generation Large Language Models 提出FeRG-LLM，利用Reason Generation LLM自动进行表格数据特征工程。 DPO direct preference optimization large language model
20 CoRanking: Collaborative Ranking with Small and Large Ranking Agents CoRanking：结合大小排序模型的协同排序框架，提升排序效率和效果。 reinforcement learning large language model
21 ToRL: Scaling Tool-Integrated RL 提出ToRL框架，通过强化学习训练LLM自主使用计算工具，显著提升数学问题求解能力。 reinforcement learning large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页