cs.CL（2024-12-12）

📊 共 24 篇论文 | 🔗 6 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (21 🔗6) 支柱二：RL算法与架构 (RL & Architecture) (2) 支柱六：视频提取与匹配 (Video Extraction) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (21 篇)

#题目一句话要点标签🔗⭐
1 Dipper: Diversity in Prompts for Producing Large Language Model Ensembles in Reasoning tasks DIPPER：通过多样化Prompt生成大语言模型推理集成，提升小模型性能 large language model
2 Filter-then-Generate: Large Language Models with Structure-Text Adapter for Knowledge Graph Completion 提出FtG：一种基于结构-文本适配器的大语言模型知识图谱补全方法 large language model ✅
3 When Text Embedding Meets Large Language Model: A Comprehensive Survey 综述：当文本嵌入遇见大语言模型，探索融合与演进 large language model
4 OG-RAG: Ontology-Grounded Retrieval-Augmented Generation For Large Language Models 提出OG-RAG，利用本体知识增强LLM在特定领域的事实推理能力。 large language model
5 Exploring Large Language Models on Cross-Cultural Values in Connection with Training Methodology 探索大型语言模型在跨文化价值观判断中的表现及与训练方法的关系 large language model
6 Foundational Large Language Models for Materials Research 提出LLaMat：材料科学领域专用大语言模型，提升材料发现与结构预测能力 large language model
7 The Impact of Copyrighted Material on Large Language Models: A Norwegian Perspective 评估版权材料对挪威语大型语言模型性能的影响 large language model
8 Forest-of-Thought: Scaling Test-Time Compute for Enhancing LLM Reasoning 提出Forest-of-Thought框架，提升LLM在复杂推理任务中的精度和效率 large language model chain-of-thought ✅
9 AgentTrek: Agent Trajectory Synthesis via Guiding Replay with Web Tutorials AgentTrek：利用Web教程引导回放合成GUI Agent轨迹数据 multimodal chain-of-thought
10 A NotSo Simple Way to Beat Simple Bench 提出迭代推理框架，提升大语言模型在逻辑连贯性和现实世界推理上的能力 large language model
11 Towards Understanding the Robustness of LLM-based Evaluations under Perturbations 探讨LLM评估在扰动下的鲁棒性问题 large language model
12 What Makes Cryptic Crosswords Challenging for LLMs? 评估大型语言模型在隐晦填字游戏中的表现并探究其挑战 large language model ✅
13 RuleArena: A Benchmark for Rule-Guided Reasoning with LLMs in Real-World Scenarios RuleArena：一个评估LLM在真实场景中规则引导推理能力的新基准 large language model ✅
14 GReaTer: Gradients over Reasoning Makes Smaller Language Models Strong Prompt Optimizers GReaTer：利用推理梯度优化小模型Prompt，提升其在复杂任务上的性能 large language model ✅
15 OpenNER 1.0: Standardized Open-Access Named Entity Recognition Datasets in 50+ Languages 提出OpenNER 1.0以解决多语言命名实体识别数据集标准化问题 large language model ✅
16 Learning to Solve Domain-Specific Calculation Problems with Knowledge-Intensive Programs Generator 提出KIPG：利用知识密集型程序生成器解决领域特定计算问题 large language model
17 Make Satire Boring Again: Reducing Stylistic Bias of Satirical Corpus by Utilizing Generative LLMs 利用生成式LLM降低讽刺语料库的文体偏见，提升讽刺检测模型的泛化能力 large language model
18 ReFF: Reinforcing Format Faithfulness in Language Models across Varied Tasks 提出ReFF框架，强化大语言模型在多任务中的格式一致性 large language model
19 ZigZagkv: Dynamic KV Cache Compression for Long-context Modeling based on Layer Uncertainty 提出ZigZagkv，利用层不确定性动态压缩长文本建模中的KV缓存，显著降低内存占用。 large language model
20 Align, Generate, Learn: A Novel Closed-Loop Framework for Cross-Lingual In-Context Learning 提出一种新型闭环框架，用于提升跨语言上下文学习的性能和泛化性。 large language model
21 GRIP: A Graph-Based Reasoning Instruction Producer 提出GRIP：一种基于图推理指令生成器，用于高效合成高质量、多样化的推理数据。 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (2 篇)

#题目一句话要点标签🔗⭐
22 From Text to Trajectory: Exploring Complex Constraint Representation and Decomposition in Safe Reinforcement Learning 提出轨迹级文本约束翻译器TTCT，解决安全强化学习中复杂文本约束的表示与分解问题 reinforcement learning zero-shot transfer
23 Reasoning-Aware Query-Focused Summarization over Multi-Table Data 提出QueryTableSummarizer++，解决多表数据推理感知查询聚焦摘要生成问题 reinforcement learning large language model

🔬 支柱六：视频提取与匹配 (Video Extraction) (1 篇)

#题目一句话要点标签🔗⭐
24 CleanComedy: Creating Friendly Humor through Generative Techniques CleanComedy：通过生成技术创建友好的幽默内容 HuMoR

⬅️ 返回 cs.CL 首页 · 🏠 返回主页