cs.CL（2025-07-07）

📊 共 30 篇论文 | 🔗 1 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (25) 支柱二：RL算法与架构 (RL & Architecture) (3) 支柱一：机器人控制 (Robot Control) (1 🔗1) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (25 篇)

#题目一句话要点标签🔗⭐
1 "Lost-in-the-Later": Framework for Quantifying Contextual Grounding in Large Language Models 提出CoPE框架，揭示LLM在上下文理解中存在的“后置信息丢失”现象 large language model chain-of-thought
2 MindFlow: Revolutionizing E-commerce Customer Support with Multimodal LLM Agents MindFlow：利用多模态LLM Agent革新电商客户支持 large language model multimodal
3 Mechanistic Indicators of Understanding in Large Language Models 通过机制可解释性，探究大语言模型中理解能力的涌现与层次 large language model
4 On the Semantics of Large Language Models 探究大型语言模型在词汇和句子层面的语义理解能力 large language model
5 Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities Gemini 2.5：通过高级推理、多模态、长上下文和新一代Agent能力突破前沿 multimodal
6 An Evaluation of Large Language Models on Text Summarization Tasks Using Prompt Engineering Techniques 通过Prompt工程评估大型语言模型在文本摘要任务上的性能，并提出句子分块策略优化长文档摘要。 large language model
7 PRIME: Large Language Model Personalization with Cognitive Dual-Memory and Personalized Thought Process PRIME：利用认知双记忆和个性化思维过程实现大语言模型个性化 large language model
8 $\textit{Grahak-Nyay:}$ Consumer Grievance Redressal through Large Language Models Grahak-Nyay：利用大型语言模型解决印度消费者权益纠纷问题。 large language model
9 Agentic Vehicles for Human-Centered Mobility 提出Agentic Vehicles (AgVs)概念，弥合自动驾驶技术与以人为本的出行需求之间的差距。 large language model multimodal
10 ArtifactsBench: Bridging the Visual-Interactive Gap in LLM Code Generation Evaluation ArtifactsBench：弥合LLM代码生成评估中视觉交互的鸿沟 large language model multimodal
11 Why We Feel What We Feel: Joint Detection of Emotions and Their Opinion Triggers in E-commerce 提出EOT-DETECT框架，联合检测电商评论中的情绪及其触发因素。 large language model chain-of-thought
12 On the Bias of Next-Token Predictors Toward Systematically Inefficient Reasoning: A Shortest-Path Case Study 研究表明，在最短路径推理任务中，基于低效推理轨迹训练的LLM泛化性更强。 large language model
13 Reason to Rote: Rethinking Memorization in Reasoning 探究LLM中记忆噪声标签与推理能力的关系，揭示良性记忆的机理 large language model
14 Controlling What You Share: Assessing Language Model Adherence to Privacy Preferences 提出基于隐私配置文件的LLM查询重写框架，提升用户数据隐私保护。 large language model
15 Evaluating Memory in LLM Agents via Incremental Multi-Turn Interactions 提出MemoryAgentBench，用于评估LLM Agent在多轮交互中的记忆能力 large language model
16 Interpretable Mnemonic Generation for Kanji Learning via Expectation-Maximization 提出基于期望最大化的可解释助记符生成方法，辅助汉字学习 large language model
17 Emergent Semantics Beyond Token Embeddings: Transformer LMs with Frozen Visual Unicode Representations 提出基于冻结视觉Unicode表征的Transformer LM，突破传统token嵌入语义限制。 large language model
18 Knowledge-Aware Self-Correction in Language Models via Structured Memory Graphs 提出基于结构化记忆图的知识感知自校正框架，提升语言模型的事实准确性 large language model
19 PhoniTale: Phonologically Grounded Mnemonic Generation for Typologically Distant Language Pairs PhoniTale：面向音系学的助记符生成，解决跨语系语言学习的词汇习得难题 large language model
20 Co-DETECT: Collaborative Discovery of Edge Cases in Text Classification Co-DETECT：结合人类专家知识与大语言模型，协同发现文本分类中的边界案例 large language model
21 Dialogue-Based Multi-Dimensional Relationship Extraction from Novels 提出基于LLM的对话式多维度关系抽取方法，用于解决小说领域的人物关系抽取难题。 large language model
22 Spec-TOD: A Specialized Instruction-Tuned LLM Framework for Efficient Task-Oriented Dialogue Systems Spec-TOD：面向任务型对话系统的高效指令调优LLM框架 large language model
23 LLMs as Architects and Critics for Multi-Source Opinion Summarization 提出M-OS-EVAL基准数据集，并探索LLM在多源意见摘要中的应用，显著提升用户参与度。 large language model
24 "This Suits You the Best": Query Focused Comparative Explainable Summarization 提出查询聚焦的比较型可解释摘要生成方法，并构建MS-Q2P数据集。 large language model
25 LOOM-Scope: a comprehensive and efficient LOng-cOntext Model evaluation framework LOOM-Scope：一个全面高效的长文本模型评估框架 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🔗⭐
26 From Fragments to Facts: A Curriculum-Driven DPO Approach for Generating Hindi News Veracity Explanations 提出基于课程学习的DPO框架，用于生成印地语新闻真伪解释 DPO direct preference optimization curriculum learning
27 SMART: Simulated Students Aligned with Item Response Theory for Question Difficulty Prediction SMART：通过模拟学生与IRT对齐，预测题目难度 DPO direct preference optimization large language model
28 Put Teacher in Student's Shoes: Cross-Distillation for Ultra-compact Model Compression Framework 提出EI-BERT框架，通过跨蒸馏实现面向边缘计算的超小型NLP模型压缩。 distillation

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
29 Response Attack: Exploiting Contextual Priming to Jailbreak Large Language Models 提出Response Attack，利用上下文启动效应破解大型语言模型 manipulation large language model ✅

🔬 支柱八：物理动画 (Physics-based Animation) (1 篇)

#题目一句话要点标签🔗⭐
30 Verified Language Processing with Hybrid Explainability: A Technical Report 提出混合可解释性框架，提升自然语言处理中全文相似度判断与逻辑推理能力 spatiotemporal

⬅️ 返回 cs.CL 首页 · 🏠 返回主页