cs.CL（2025-04-24）

📊 共 23 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) (20 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (3)

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

#题目一句话要点标签🔗⭐
1 An Empirical Study on Prompt Compression for Large Language Models 研究提示压缩方法以降低大语言模型计算成本并提升长文本处理性能。 large language model multimodal ✅
2 Bridging Cognition and Emotion: Empathy-Driven Multimodal Misinformation Detection 提出双重角度共情框架DAE，用于多模态信息错误检测。 large language model multimodal
3 M-MRE: Extending the Mutual Reinforcement Effect to Multimodal Information Extraction 首次将互增强效应扩展到多模态信息抽取，提出M-MRE任务与Prompt Format Adapter。 multimodal
4 Energy Considerations of Large Language Model Inference and Efficiency Optimizations 针对LLM推理能耗问题，提出工作负载建模方法并优化推理效率 large language model
5 Multilingual Performance Biases of Large Language Models in Education 评估大型语言模型在多语言教育场景下的性能偏差，揭示低资源语言的性能瓶颈。 large language model
6 Evaluating Intermediate Reasoning of Code-Assisted Large Language Models for Mathematics 深入评估代码辅助大语言模型在数学推理中的中间推理过程 large language model
7 PatientDx: Merging Large Language Models for Protecting Data-Privacy in Healthcare PatientDx：一种用于医疗数据隐私保护的LLM模型融合框架 large language model
8 Aleph-Alpha-GermanWeb: Improving German-language LLM pre-training with model-based data curation and synthetic data generation Aleph-Alpha-GermanWeb：通过模型驱动的数据筛选与合成数据生成，提升德语LLM预训练效果。 large language model
9 Assessing the Potential of Generative Agents in Crowdsourced Fact-Checking 利用生成式Agent模拟众包事实核查，提升效率与降低偏见 large language model
10 Ensemble Bayesian Inference: Leveraging Small Language Models to Achieve LLM-level Accuracy in Profile Matching Tasks 提出Ensemble Bayesian Inference，利用小语言模型集合在用户画像匹配任务中达到LLM级别精度 large language model
11 Beyond Public Access in LLM Pre-Training Data 利用DE-COP推断，揭示GPT-4o对版权书籍内容的更强记忆能力 large language model
12 HalluLens: LLM Hallucination Benchmark 提出HalluLens基准以解决大语言模型幻觉问题 large language model
13 LiveLongBench: Tackling Long-Context Understanding for Spoken Texts from Live Streams 提出LiveLongBench，用于评估LLM在直播场景下长文本理解能力 large language model ✅
14 FLUKE: A Linguistically-Driven and Task-Agnostic Framework for Robustness Evaluation FLUKE：一种语言驱动且任务无关的鲁棒性评估框架 large language model
15 Toward a Human-Centered Evaluation Framework for Trustworthy LLM-Powered GUI Agents 提出面向LLM驱动GUI代理的人本评估框架，关注隐私与安全风险 large language model
16 Conversational Assistants to support Heart Failure Patients: comparing a Neurosymbolic Architecture with ChatGPT 对比神经符号架构与ChatGPT在心力衰竭患者辅助对话系统中的应用 large language model
17 DeepDistill: Enhancing LLM Reasoning Capabilities via Large-Scale Difficulty-Graded Data Training DeepDistill：通过大规模难度分级数据训练提升LLM推理能力 large language model ✅
18 CoheMark: A Novel Sentence-Level Watermark for Enhanced Text Quality CoheMark：一种增强文本质量的新型句子级水印方法 large language model
19 A RAG-Based Multi-Agent LLM System for Natural Hazard Resilience and Adaptation 提出基于RAG的多Agent LLM系统WildfireGPT，用于自然灾害适应和决策支持 large language model
20 Paper2Code: Automating Code Generation from Scientific Papers in Machine Learning Paper2Code：提出一种多智能体LLM框架，自动将机器学习论文转化为可执行代码。 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🔗⭐
21 Unified attacks to large language model watermarks: spoofing and scrubbing in unauthorized knowledge distillation 提出CDG-KD框架，实现对大语言模型水印的伪造与擦除攻击，对抗非授权知识蒸馏。 distillation large language model
22 Does Knowledge Distillation Matter for Large Language Model based Bundle Generation? 提出知识蒸馏框架，提升大语言模型在Bundle生成任务中的效率与性能。 distillation large language model
23 JurisCTC: Enhancing Legal Judgment Prediction via Cross-Domain Transfer and Contrastive Learning JurisCTC：通过跨领域迁移学习和对比学习增强法律判决预测 contrastive learning large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页