cs.CL（2024-10-04）

📊 共 47 篇论文 | 🔗 9 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (41 🔗8) 支柱二：RL算法与架构 (RL & Architecture) (6 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (41 篇)

#题目一句话要点标签🔗⭐
1 RIPPLECOT: Amplifying Ripple Effect of Knowledge Editing in Language Models via Chain-of-Thought In-Context Learning 提出RippleCOT，通过思维链上下文学习增强语言模型知识编辑的涟漪效应。 large language model chain-of-thought
2 CommonIT: Commonality-Aware Instruction Tuning for Large Language Models via Data Partitions 提出CommonIT以解决大语言模型指令调优问题 large language model instruction following ✅
3 Zero-Shot Fact Verification via Natural Logic and Large Language Models 提出一种基于自然逻辑和大型语言模型的零样本事实核查方法。 large language model zero-shot transfer
4 What do Large Language Models Need for Machine Translation Evaluation? 研究大型语言模型在机器翻译评估中的信息需求与提示策略 large language model chain-of-thought
5 PersoBench: Benchmarking Personalized Response Generation in Large Language Models PersoBench：用于评估大语言模型个性化回复生成能力的新基准 large language model chain-of-thought ✅
6 Using Prompts to Guide Large Language Models in Imitating a Real Person's Language Style 利用提示工程引导大语言模型模仿个人语言风格，提升对话AI个性化 large language model
7 Towards Linguistically-Aware and Language-Independent Tokenization for Large Language Models (LLMs) 针对LLM的语言感知和语言无关的分词方法研究，提升低资源语言支持 large language model
8 Still Not Quite There! Evaluating Large Language Models for Comorbid Mental Health Diagnosis 提出ANGST基准，评估大语言模型在抑郁-焦虑共病诊断中的能力 large language model
9 SWE-bench Multimodal: Do AI Systems Generalize to Visual Software Domains? 提出SWE-bench Multimodal，评估AI系统在视觉软件领域的泛化能力。 multimodal
10 Output Scouting: Auditing Large Language Models for Catastrophic Responses 提出Output Scouting方法，高效审计大语言模型中的灾难性输出 large language model ✅
11 One2set + Large Language Model: Best Partners for Keyphrase Generation 提出One2set+LLM框架，通过生成-选择策略提升关键短语生成效果 large language model
12 Generating bilingual example sentences with large language models as lexicography assistants 利用大型语言模型生成双语例句，辅助词典编纂工作 large language model
13 PersonalSum: A User-Subjective Guided Personalized Summarization Dataset for Large Language Models PersonalSum：一个用户主观引导的个性化摘要数据集，用于评估大型语言模型 large language model
14 A Large Language Model-based Framework for Semi-Structured Tender Document Retrieval-Augmented Generation 提出基于大语言模型的招标文档检索增强生成框架，提升专业文档生成质量 large language model
15 Steering Large Language Models between Code Execution and Textual Reasoning 提出三种方法，提升大型语言模型在代码执行与文本推理间的引导能力 large language model ✅
16 CliMedBench: A Large-Scale Chinese Benchmark for Evaluating Medical Large Language Models in Clinical Scenarios 提出CliMedBench：大规模中文医疗LLM评测基准，聚焦临床场景 large language model
17 Context and System Fusion in Post-ASR Emotion Recognition with Large Language Models 利用大语言模型融合上下文和多系统输出，提升ASR后情感识别准确率 large language model
18 Consultation on Industrial Machine Faults with Large language Models 提出基于多轮提示的大语言模型方法，用于工业机器故障诊断 large language model
19 Kiss up, Kick down: Exploring Behavioral Changes in Multi-modal Large Language Models with Assigned Visual Personas 首个探索视觉角色对多模态大语言模型行为影响的研究 large language model
20 Autoregressive Large Language Models are Computationally Universal 证明自回归大语言模型在计算上具有通用性，无需外部干预。 large language model
21 Self-Powered LLM Modality Expansion for Large Speech-Text Models 提出自驱动LLM模态扩展方法，解决语音-文本大模型中的语音锚定偏差问题 large language model multimodal instruction following ✅
22 Searching for Best Practices in Medical Transcription with Large Language Model 利用大型语言模型提升医疗转录准确率，尤其针对印度口音 large language model
23 ActPlan-1K: Benchmarking the Procedural Planning Ability of Visual Language Models in Household Activities ActPlan-1K：用于评估视觉语言模型在家庭活动中程序规划能力的基准 embodied AI large language model
24 SAG: Style-Aligned Article Generation via Model Collaboration 提出SAG：一种基于模型协作的风格对齐文章生成方法，显著提升生成质量。 large language model instruction following
25 LLM-TOPLA: Efficient LLM Ensemble by Maximising Diversity LLM-TOPLA：通过最大化多样性实现高效的大语言模型集成 large language model ✅
26 Detecting Machine-Generated Long-Form Content with Latent-Space Variables 提出潜在空间变量模型以解决机器生成长文本检测问题 large language model
27 RAFT: Realistic Attacks to Fool Text Detectors RAFT：提出一种针对LLM检测器的逼真黑盒攻击方法，提升攻击的隐蔽性和有效性。 large language model
28 Auto-GDA: Automatic Domain Adaptation for Efficient Grounding Verification in Retrieval-Augmented Generation 提出Auto-GDA，通过自动领域自适应提升RAG中 grounding verification 的效率。 large language model
29 Mixture of Attentions For Speculative Decoding 提出混合注意力机制用于推测解码，提升单设备和客户端-服务器场景下的解码速度和精度。 large language model
30 KidLM: Advancing Language Models for Children -- Early Insights and Future Directions KidLM：面向儿童的语言模型，通过定制数据和训练目标提升性能。 large language model
31 Scalable Frame-based Construction of Sociocultural NormBases for Socially-Aware Dialogues 提出一种可扩展的框架，利用大型语言模型构建用于社交对话的社会文化规范库。 large language model
32 ORAssistant: A Custom RAG-based Conversational Assistant for OpenROAD 提出ORAssistant，一个基于RAG的OpenROAD定制对话助手 large language model
33 Re-examining Sexism and Misogyny Classification with Annotator Attitudes 通过考察标注者态度，重新审视性别歧视和厌女症分类问题 large language model
34 Towards Reproducible LLM Evaluation: Quantifying Uncertainty in LLM Benchmark Scores 量化LLM基准测试不确定性，提升评估可复现性 large language model
35 Generating Equivalent Representations of Code By A Self-Reflection Approach 提出一种自反思方法，利用大语言模型自动生成代码的等价表示 large language model
36 Beyond Film Subtitles: Is YouTube the Best Approximation of Spoken Vocabulary? 利用YouTube字幕构建高质量词频资源，提升心理语言学和词汇复杂度预测任务性能 large language model ✅
37 Showing LLM-Generated Code Selectively Based on Confidence of LLMs HonestCoder：基于置信度选择性展示LLM生成代码，提升开发者效率并降低安全风险 large language model
38 Can Watermarked LLMs be Identified by Users via Crafted Prompts? 提出Water-Probe算法以识别水印LLM的隐蔽性问题 large language model
39 X-ALMA: Plug & Play Modules and Adaptive Rejection for Quality Translation at Scale X-ALMA：通过即插即用模块和自适应拒绝优化，实现大规模高质量翻译 large language model
40 UNComp: Can Matrix Entropy Uncover Sparsity? -- A Compressor Design from an Uncertainty-Aware Perspective 提出UNComp，利用矩阵熵指导LLM的KV缓存压缩，提升长文本推理效率。 large language model ✅
41 Enhancing Short-Text Topic Modeling with LLM-Driven Context Expansion and Prefix-Tuned VAEs 提出LLM驱动的上下文扩展与Prefix-Tuned VAEs方法，提升短文本主题建模效果。 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (6 篇)

#题目一句话要点标签🔗⭐
42 Crafting Narrative Closures: Zero-Shot Learning with SSM Mamba for Short Story Ending Generation 提出基于SSM Mamba的零样本短故事结局生成方法，助力创意写作。 Mamba SSM
43 Learning from Committee: Reasoning Distillation from a Mixture of Teachers with Peer-Review 提出FAIR：通过委员会推理蒸馏，利用同伴互评机制提升小模型推理能力 distillation large language model ✅
44 ToolGen: Unified Tool Retrieval and Calling via Generation ToolGen：通过生成式方法统一工具检索与调用，提升LLM自主任务执行能力。 reinforcement learning large language model chain-of-thought
45 Deliberate Reasoning in Language Models as Structure-Aware Planning with an Accurate World Model 提出SWAP框架，通过结构化知识和规划提升语言模型在复杂推理任务中的性能。 world model chain-of-thought
46 Aligning LLMs with Individual Preferences via Interaction 通过交互对齐LLM与个体偏好：提出ALOE基准与个性化对齐方法 reinforcement learning large language model
47 Margin Matching Preference Optimization: Enhanced Model Alignment with Granular Feedback 提出边际匹配偏好优化以解决模型对齐问题 reinforcement learning large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页