cs.CL（2024-11-12）

📊 共 17 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (14 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (3 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (14 篇)

#题目一句话要点标签🔗⭐
1 Derivational Morphology Reveals Analogical Generalization in Large Language Models 提出类比过程以揭示大型语言模型的语言推广机制 large language model
2 Large Language Models Can Self-Improve in Long-context Reasoning 提出自我改进方法以解决长上下文推理问题 large language model
3 Verbosity $\neq$ Veracity: Demystify Verbosity Compensation Behavior of Large Language Models 揭示大语言模型的冗余补偿行为，并提出简单有效的缓解方法 large language model ✅
4 Mitigating Bias in Queer Representation within Large Language Models: A Collaborative Agent Approach 提出一种协同Agent框架，用于缓解大型语言模型中对酷儿群体表征的偏见。 large language model
5 Large Language Models as Neurolinguistic Subjects: Discrepancy between Performance and Competence 提出神经语言学方法以评估大型语言模型的语言能力 large language model
6 DecoPrompt : Decoding Prompts Reduces Hallucinations when Large Language Models Meet False Premises DecoPrompt：通过解码提示来减少大语言模型在错误前提下的幻觉 large language model
7 ExpressivityArena: Can LLMs Express Information Implicitly? 提出ExpressivityArena，用于评估LLM的隐式信息表达能力 large language model
8 Trustful LLMs: Customizing and Grounding Text Generation with Knowledge Bases and Dual Decoders 提出基于知识库和双解码器的可信LLM，提升生成内容的事实性和完整性 large language model
9 Chain Association-based Attacking and Shielding Natural Language Processing Systems 提出基于链式联想的对抗攻击方法，针对自然语言处理系统 large language model
10 Problem-Oriented Segmentation and Retrieval: Case Study on Tutoring Conversations 提出问题导向的分割与检索(POSR)框架，应用于辅导对话结构化分析。 large language model
11 Beyond the Safety Bundle: Auditing the Helpful and Harmless Dataset 审核Helpful and Harmless数据集，揭示其安全性和公平性问题 large language model
12 SHARP: Unlocking Interactive Hallucination via Stance Transfer in Role-Playing LLMs 提出SHARP以解决角色扮演LLM中的互动幻觉问题 large language model
13 Rapid Response: Mitigating LLM Jailbreaks with a Few Examples 提出快速响应方法，仅用少量样本显著缓解LLM越狱攻击 large language model
14 Efficient and Accurate Prompt Optimization: the Benefit of Memory in Exemplar-Guided Reflection 提出ERM框架，利用记忆机制优化Prompt，提升大语言模型生成质量。 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🔗⭐
15 Knowledge-Augmented Multimodal Clinical Rationale Generation for Disease Diagnosis with Small Language Models 提出ClinRaGen，通过知识增强的小语言模型进行多模态临床诊断和推理生成。 distillation large language model multimodal
16 Query Optimization for Parametric Knowledge Refinement in Retrieval-Augmented Large Language Models 提出ERRR框架，通过优化查询提升检索增强大语言模型中的参数知识利用率。 distillation large language model
17 Prompt-enhanced Network for Hateful Meme Classification 提出Pen：一种提示增强网络，用于提升仇恨模因分类的准确性和泛化性 contrastive learning multimodal ✅

⬅️ 返回 cs.CL 首页 · 🏠 返回主页