cs.CL（2025-02-13）

📊 共 33 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (30 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (3)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (30 篇)

#题目一句话要点标签🔗⭐
1 SQuARE: Sequential Question Answering Reasoning Engine for Enhanced Chain-of-Thought in Large Language Models SQuARE：通过自问自答增强LLM链式思考推理能力 large language model chain-of-thought ✅
2 Large Language Models and Provenance Metadata for Determining the Relevance of Images and Videos in News Stories 提出基于大语言模型和溯源元数据的多模态信息可信度评估方法 large language model multimodal
3 SelfCite: Self-Supervised Alignment for Context Attribution in Large Language Models SelfCite：一种自监督对齐方法，用于大语言模型中的上下文归因。 large language model ✅
4 Zero-shot generation of synthetic neurosurgical data with large language models 利用大型语言模型零样本生成合成神经外科数据，解决数据稀缺问题。 large language model
5 Quantifying depressive mental states with large language models 利用大型语言模型量化抑郁症心理状态，揭示数据需求与概念对齐的界限。 large language model
6 Trust at Your Own Peril: A Mixed Methods Exploration of the Ability of Large Language Models to Generate Expert-Like Systems Engineering Artifacts and a Characterization of Failure Modes 评估大语言模型生成系统工程制品能力，揭示其潜在失效模式及风险 large language model
7 Mind What You Ask For: Emotional and Rational Faces of Persuasion by Large Language Models 研究大型语言模型在情感和理性提示下的说服策略，揭示潜在的误导风险。 large language model
8 The Widespread Adoption of Large Language Model-Assisted Writing Across Society 分析大型语言模型辅助写作在社会各领域的普及程度与应用模式 large language model
9 Structured Convergence in Large Language Model Representations via Hierarchical Latent Space Folding 提出层级潜在空间折叠方法，提升大语言模型表征的结构性和计算效率。 large language model
10 FoNE: Precise Single-Token Number Embeddings via Fourier Features FoNE：通过傅里叶特征实现精确的单token数值嵌入，提升LLM数值计算能力。 large language model ✅
11 Human-LLM Coevolution: Evidence from Academic Writing 通过分析arXiv摘要，揭示人类作者与LLM在学术写作中的协同进化现象 large language model
12 Hope vs. Hate: Understanding User Interactions with LGBTQ+ News Content in Mainstream US News Media through the Lens of Hope Speech 通过希望言论分析用户与美国主流媒体LGBTQ+新闻内容的互动 large language model
13 Beyond the Singular: Revealing the Value of Multiple Generations in Benchmark Evaluation 提出分层统计模型以提升大语言模型基准评估的准确性 large language model
14 InfiniteHiP: Extending Language Model Context Up to 3 Million Tokens on a Single GPU InfiniteHiP：在单GPU上扩展语言模型上下文至300万tokens large language model
15 Mind the Gap! Choice Independence in Using Multilingual LLMs for Persuasive Co-Writing Tasks in Different Languages 多语言LLM用于广告文案写作时，用户选择独立性受损，影响捐赠意愿 large language model
16 Improve LLM-based Automatic Essay Scoring with Linguistic Features 融合语言特征提升LLM自动作文评分性能 large language model
17 SparQLe: Speech Queries to Text Translation Through LLMs SparQLe：提出一种基于LLM的语音查询到文本翻译方法 large language model
18 Matina: A Large-Scale 73B Token Persian Text Corpus 发布大规模波斯语文本语料库Matina，促进波斯语NLP模型发展 large language model
19 CoSER: A Comprehensive Literary Dataset and Framework for Training and Evaluating LLM Role-Playing and Persona Simulation CoSER：构建综合文学数据集与框架，用于训练和评估LLM的角色扮演和人物模拟能力 large language model
20 Adapting Language-Specific LLMs to a Reasoning Model in One Day via Model Merging -- An Open Recipe 通过模型合并，一天内将特定语言LLM适配到推理模型：一个开放方案 large language model
21 The Hidden Dimensions of LLM Alignment: A Multi-Dimensional Analysis of Orthogonal Safety Directions 揭示LLM对齐的隐藏维度：正交安全方向的多维分析 large language model ✅
22 Are Smarter LLMs Safer? Exploring Safety-Reasoning Trade-offs in Prompting and Fine-Tuning 研究提示与微调中推理能力与安全性之间的权衡 large language model
23 Towards Automated Fact-Checking of Real-World Claims: Exploring Task Formulation and Assessment with LLMs 利用LLM进行自动化事实核查：任务形式与评估方法研究 large language model
24 Communication is All You Need: Persuasion Dataset Construction via Multi-LLM Communication 提出基于多LLM通信的框架，自动生成高质量、多样化的说服性对话数据集。 large language model
25 INJONGO: A Multicultural Intent Detection and Slot-filling Dataset for 16 African Languages 提出Injongo：一个面向16种非洲语言的多文化意图检测与槽填充数据集 large language model
26 Logical forms complement probability in understanding language model (and human) performance 探究逻辑形式在理解语言模型和人类表现中的互补作用 large language model
27 A Judge-free LLM Open-ended Generation Benchmark Based on the Distributional Hypothesis 提出基于分布假设的无监督LLM开放生成评估基准 large language model
28 Thinking beyond the anthropomorphic paradigm benefits LLM research 挑战拟人化范式：为大语言模型研究开辟新路径 large language model
29 Can Vision-Language Models Infer Speaker's Ignorance? The Role of Visual and Linguistic Cues 研究视觉语言模型能否基于视觉和语言线索推断说话者的无知含义 multimodal
30 Enhancing RAG with Active Learning on Conversation Records: Reject Incapables and Answer Capables 提出AL4RAG，通过主动学习提升RAG在对话记录上的抗幻觉能力 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🔗⭐
31 Statistical Coherence Alignment for Large Language Model Representation Learning Through Tensor Field Convergence 提出统计一致性对齐方法，通过张量场收敛提升大语言模型表征学习效果 representation learning large language model
32 RefineCoder: Iterative Improving of Large Language Models via Adaptive Critique Refinement for Code Generation RefineCoder：通过自适应评价改进迭代提升大语言模型的代码生成能力 distillation large language model
33 Typhoon T1: An Open Thai Reasoning Model Typhoon T1：一个开放的泰语推理模型，促进低资源语言的推理研究。 reinforcement learning large language model chain-of-thought

⬅️ 返回 cs.CL 首页 · 🏠 返回主页