cs.CL（2024-12-13）

📊 共 31 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) (26 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (5)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (26 篇)

#题目一句话要点标签🔗⭐
1 MERaLiON-AudioLLM: Bridging Audio and Language with Large Language Models MERaLiON-AudioLLM：针对新加坡多语环境的语音-文本大型语言模型 large language model multimodal
2 Still "Talking About Large Language Models": Some Clarifications 澄清对大型语言模型讨论的误解，强调语言使用而非形而上学 large language model
3 On Adversarial Robustness and Out-of-Distribution Robustness of Large Language Models 研究大型语言模型在对抗攻击和分布外数据上的鲁棒性关联与迁移性。 large language model
4 Targeted Angular Reversal of Weights (TARS) for Knowledge Removal in Large Language Models 提出TARS方法，通过定向反转权重有效移除大语言模型中的特定知识。 large language model
5 Small Language Model as Data Prospector for Large Language Model SuperNUGGETS：利用小语言模型高效筛选高质量指令数据，提升大语言模型微调效果。 large language model
6 Enhancing Nursing and Elderly Care with Large Language Models: An AI-Driven Framework 提出基于LLM的AI驱动框架，增强护理和老年人照护能力 large language model
7 Semi-IIN: Semi-supervised Intra-inter modal Interaction Learning Network for Multimodal Sentiment Analysis 提出Semi-IIN，利用半监督学习和动态交互选择解决多模态情感分析标注成本高和交互选择难题。 multimodal ✅
8 WHAT-IF: Exploring Branching Narratives by Meta-Prompting Large Language Models 提出WHAT-IF系统以探索分支叙事的交互式小说 large language model ✅
9 Benchmarking Linguistic Diversity of Large Language Models 提出评估大型语言模型语言多样性的框架 large language model
10 AMuSeD: An Attentive Deep Neural Network for Multimodal Sarcasm Detection Incorporating Bi-modal Data Augmentation AMuSeD：融合双模态数据增强的注意力深度网络用于多模态讽刺检测 multimodal
11 Large Language Models for Persian $ \leftrightarrow $ English Idiom Translation 利用大型语言模型进行波斯语-英语习语翻译研究 large language model
12 Human-Like Embodied AI Interviewer: Employing Android ERICA in Real International Conference 提出类人具身AI面试官ERICA，首次应用于国际会议SIGDIAL 2024 embodied AI
13 Enhancing the Reasoning Capabilities of Small Language Models via Solution Guidance Fine-Tuning 提出Solution Guidance Fine-Tuning，提升小语言模型推理能力 large language model chain-of-thought
14 A Grounded Typology of Word Classes 提出一种基于多模态语言模型的词类语义内容度量方法，用于跨语言词类类型学研究。 multimodal
15 Low-Rank Adaptation with Task-Relevant Feature Enhancement for Fine-tuning Language Models 提出LoRATRF，通过任务相关特征增强提升LoRA微调语言模型性能 large language model
16 Too Big to Fool: Resisting Deception in Language Models 研究表明，更大规模语言模型更能抵抗提示中的欺骗信息 large language model
17 Efficient Continual Pre-training of LLMs for Low-resource Languages 提出高效的LLM持续预训练方法，降低低资源语言的训练成本并提升性能。 large language model
18 ROUTE: Robust Multitask Tuning and Collaboration for Text-to-SQL 提出ROUTE方法，通过多任务调优和协作提升开源LLM在Text-to-SQL任务上的性能。 large language model
19 GAOKAO-Eval: Does high scores truly reflect strong capabilities in LLMs? 提出GAOKAO-Eval以评估LLMs真实能力 large language model
20 ChainStream: An LLM-based Framework for Unified Synthetic Sensing ChainStream：基于LLM的统一合成感知框架，简化应用开发并提升数据透明度。 large language model ✅
21 Modeling Story Expectations to Understand Engagement: A Generative Framework Using LLMs 利用LLM建模故事期望以理解用户参与度：一种生成式框架 large language model
22 AutoPatent: A Multi-Agent Framework for Automatic Patent Generation 提出AutoPatent框架以解决自动专利生成问题 large language model ✅
23 One world, one opinion? The superstar effect in LLM responses 揭示LLM中的“超级明星效应”：语言模型意见的全球知识窄化风险 large language model
24 Retrieval-Augmented Semantic Parsing: Improving Generalization with Lexical Knowledge 提出检索增强语义解析(RASP)，利用外部知识提升开放域语义解析泛化能力。 large language model
25 ASLoRA: Adaptive Sharing Low-Rank Adaptation Across Layers ASLoRA：提出自适应跨层共享低秩适配方法，提升大模型微调效率。 large language model
26 On the Limit of Language Models as Planning Formalizers 评估大语言模型作为规划形式化器的能力，并分析自然语言描述对性能的影响 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (5 篇)

#题目一句话要点标签🔗⭐
27 MPPO: Multi Pair-wise Preference Optimization for LLMs with Arbitrary Negative Samples MPPO：面向任意负样本的大语言模型多重成对偏好优化 reinforcement learning PPO RLHF
28 LLM Distillation for Efficient Few-Shot Multiple Choice Question Answering 提出基于LLM蒸馏的少样本选择题问答高效方法 distillation large language model
29 ScaleOT: Privacy-utility-scalable Offsite-tuning with Dynamic LayerReplace and Selective Rank Compression ScaleOT：一种隐私-效用可扩展的异地调优框架，用于保护大语言模型。 reinforcement learning distillation large language model
30 Label-template based Few-Shot Text Classification with Contrastive Learning 提出基于标签模板和对比学习的小样本文本分类框架 contrastive learning
31 Reasoner Outperforms: Generative Stance Detection with Rationalization for Social Media 提出基于推理的生成式立场检测方法，提升小模型在社交媒体分析中的性能与可解释性。 distillation large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页