cs.CL（2024-10-25）

📊 共 21 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (17 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (4)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (17 篇)

#题目一句话要点标签🔗⭐
1 Improving Multimodal Large Language Models Using Continual Learning 利用持续学习提升多模态大语言模型性能，缓解语言能力退化 large language model multimodal ✅
2 Revealing and Reducing Gender Biases in Vision and Language Assistants (VLAs) 揭示并减少视觉语言助手(VLA)中的性别偏见 VLA large language model multimodal ✅
3 Graph Linearization Methods for Reasoning on Graphs with Large Language Models 提出基于图线性化的方法，利用大语言模型进行图推理。 large language model multimodal
4 Counting Ability of Large Language Models and Impact of Tokenization 研究揭示分词策略对大语言模型计数能力的影响，并提出改进方向 large language model chain-of-thought
5 Layer by Layer: Uncovering Where Multi-Task Learning Happens in Instruction-Tuned Large Language Models 揭示指令调优LLM中多任务学习发生位置，探究任务特定知识的编码方式。 large language model
6 Intelligent Understanding of Large Language Models in Traditional Chinese Medicine Based on Prompt Engineering Framework 提出TCM-Prompt框架，提升大语言模型在中医药领域的理解能力 large language model
7 Investigating the Role of Prompting and External Tools in Hallucination Rates of Large Language Models 研究提示工程与外部工具对大语言模型幻觉率的影响，发现简单提示策略更有效。 large language model
8 KAHANI: Culturally-Nuanced Visual Storytelling Tool for Non-Western Cultures KAHANI：为非西方文化打造的、具有文化细微差别的视觉故事生成工具 large language model chain-of-thought
9 Less is More: Extreme Gradient Boost Rank-1 Adaption for Efficient Finetuning of LLMs 提出XGBLoRA，通过梯度提升Rank-1 LoRA实现LLM高效微调 large language model
10 Two are better than one: Context window extension with multi-grained self-injection 提出SharedLLM，通过多粒度自注入扩展LLM上下文窗口，降低长文本处理成本。 large language model
11 Interleaving Text and Number Embeddings to Solve Mathemathics Problems 提出交错文本与数字嵌入方法，提升LLM解决数学问题的能力 large language model
12 Developing a Tutoring Dialog Dataset to Optimize LLMs for Educational Use 开发辅导对话数据集以优化LLM在教育领域的应用 large language model
13 Do Discrete Self-Supervised Representations of Speech Capture Tone Distinctions? 离散语音自监督表征在声调语言中损失声调信息 foundation model
14 ChunkRAG: Novel LLM-Chunk Filtering Method for RAG Systems ChunkRAG：提出一种新颖的LLM驱动的RAG系统Chunk过滤方法，提升事实准确性。 large language model
15 Introducing MAPO: Momentum-Aided Gradient Descent Prompt Optimization MAPO：动量辅助梯度下降提示优化，提升大语言模型提示工程效率 large language model
16 A Debate-Driven Experiment on LLM Hallucinations and Accuracy 基于辩论驱动实验探究LLM幻觉与准确性，提升模型鲁棒性 large language model
17 AgentSense: Benchmarking Social Intelligence of Language Agents through Interactive Scenarios AgentSense：通过交互式场景评估语言智能体的社会智能 large language model ✅

🔬 支柱二：RL算法与架构 (RL & Architecture) (4 篇)

#题目一句话要点标签🔗⭐
18 SWITCH: Studying with Teacher for Knowledge Distillation of Large Language Models 提出SWITCH，通过教师模型干预解决大语言模型知识蒸馏中的长序列偏差问题。 distillation large language model instruction following
19 2D-DPO: Scaling Direct Preference Optimization with 2-Dimensional Supervision 提出2D-DPO框架，利用二维监督信号提升大语言模型与人类偏好对齐效果 DPO direct preference optimization large language model
20 OpenWebVoyager: Building Multimodal Web Agents via Iterative Real-World Exploration, Feedback and Optimization OpenWebVoyager：通过迭代式真实网络探索、反馈与优化构建多模态Web Agent imitation learning multimodal
21 ShifCon: Enhancing Non-Dominant Language Capabilities with a Shift-based Multilingual Contrastive Framework ShifCon：基于Shift的多语言对比学习框架，提升非优势语言大模型能力 contrastive learning large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页