cs.CL（2024-06-26）

📊 共 46 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) (44 🔗8) 支柱七：动作重定向 (Motion Retargeting) (1) 支柱二：RL算法与架构 (RL & Architecture) (1 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (44 篇)

#题目一句话要点标签🔗⭐
1 CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs CharXiv：揭示多模态LLM在真实图表理解中的差距 large language model multimodal
2 Role-Play Zero-Shot Prompting with Large Language Models for Open-Domain Human-Machine Conversation 提出基于角色扮演的零样本提示方法，提升大语言模型在开放域人机对话中的表现 large language model instruction following
3 S3: A Simple Strong Sample-effective Multimodal Dialog System 提出S3模型，一种简单高效的多模态对话系统，在MMMU和AI Journey Contest 2023上取得领先成果。 large language model multimodal
4 LLM-Driven Multimodal Opinion Expression Identification 提出基于LLM的多模态情感表达识别方法STOEI，提升语音助手和抑郁症诊断等应用的情感理解能力。 large language model multimodal
5 A Closer Look into Mixture-of-Experts in Large Language Models 深入研究大型语言模型中的混合专家（MoE）机制 large language model ✅
6 ResumeAtlas: Revisiting Resume Classification with Large-Scale Datasets and Large Language Models ResumeAtlas：利用大规模数据集和大型语言模型改进简历分类 large language model
7 AdaZeta: Adaptive Zeroth-Order Tensor-Train Adaption for Memory-Efficient Large Language Models Fine-Tuning 提出AdaZeta框架，提升MeZO方法在大语言模型微调中的性能和收敛性 large language model
8 Cascading Large Language Models for Salient Event Graph Generation 提出CALLMSAE框架，利用级联大语言模型生成文档的显著事件图，无需人工标注。 large language model
9 PaCoST: Paired Confidence Significance Testing for Benchmark Contamination Detection in Large Language Models 提出PaCoST，通过置信度显著性检验检测大语言模型中的基准污染问题 large language model
10 MathOdyssey: Benchmarking Mathematical Problem-Solving Skills in Large Language Models Using Odyssey Math Data 提出MathOdyssey数据集，用于评估大型语言模型在数学问题求解中的能力。 large language model
11 Zero-shot prompt-based classification: topic labeling in times of foundation models in German Tweets 利用零样本提示学习，解决德语推特文本的主题标注问题 foundation model
12 Enhancing Data Privacy in Large Language Models through Private Association Editing 提出私有化关联编辑(PAE)方法，无需重训练即可增强LLM的数据隐私保护。 large language model
13 Improving Entity Recognition Using Ensembles of Deep Learning and Fine-tuned Large Language Models: A Case Study on Adverse Event Extraction from Multiple Sources 通过深度学习与微调大语言模型集成，提升实体识别效果：以多源不良事件抽取为例 large language model
14 PharmaGPT: Domain-Specific Large Language Models for Bio-Pharmaceutical and Chemistry PharmaGPT：面向生物制药和化学领域的领域特定大语言模型 large language model
15 Catching Chameleons: Detecting Evolving Disinformation Generated using Large Language Models 提出DELD模型，解决大语言模型生成的不实信息持续演变带来的检测难题 large language model
16 Explicit Diversity Conditions for Effective Question Answer Generation with Large Language Models 提出显式多样性条件，提升大型语言模型生成问题答案对的质量与多样性 large language model
17 Re-Ranking Step by Step: Investigating Pre-Filtering for Re-Ranking with Large Language Models 提出基于预过滤的重排序方法，提升小模型在大语言模型重排序中的竞争力 large language model
18 Assessing "Implicit" Retrieval Robustness of Large Language Models 评估大语言模型在检索增强生成中的“隐式”检索鲁棒性 large language model
19 Octo-planner: On-device Language Model for Planner-Action Agents 提出Octo-planner，一种基于端侧语言模型的规划-行动智能体框架 Octo ✅
20 BADGE: BADminton report Generation and Evaluation with LLM BADGE：利用大型语言模型自动生成和评估羽毛球比赛报告 large language model chain-of-thought ✅
21 SEED: Accelerating Reasoning Tree Construction via Scheduled Speculative Decoding SEED：通过调度推测解码加速推理树构建 large language model chain-of-thought
22 MATE: Meet At The Embedding -- Connecting Images with Long Texts 提出MATE：通过嵌入空间对齐，连接图像与长文本 large language model
23 JailbreakZoo: Survey, Landscapes, and Horizons in Jailbreaking Large Language and Vision-Language Models JailbreakZoo：大型语言和视觉语言模型越狱攻击的综述、格局与展望 large language model
24 Psychological Profiling in Cybersecurity: A Look at LLMs and Psycholinguistic Features 利用LLM和心理语言特征进行网络安全中的心理画像分析 large language model
25 Understand What LLM Needs: Dual Preference Alignment for Retrieval-Augmented Generation DPA-RAG：通过双重偏好对齐增强检索增强生成，缓解大语言模型的幻觉问题。 large language model ✅
26 Towards Compositionality in Concept Learning 提出CCE方法，旨在提升概念学习中概念表示的组合性，从而提高模型的可解释性和下游任务性能。 foundation model ✅
27 Symbolic Learning Enables Self-Evolving Agents 提出Agent Symbolic Learning框架，使语言Agent具备数据驱动的自主进化能力 large language model
28 Do LLMs dream of elephants (when told not to)? Latent concept association and associative memory in transformers 研究发现LLM具有联想记忆特性，易受上下文操纵，并从理论上分析了Transformer的记忆机制。 large language model
29 IRCAN: Mitigating Knowledge Conflicts in LLM Generation via Identifying and Reweighting Context-Aware Neurons IRCAN：通过识别和重加权上下文感知神经元缓解LLM生成中的知识冲突 large language model ✅
30 AI-native Memory: A Pathway from LLMs Towards AGI 提出AI原生记忆，探索从LLM通往AGI的路径 large language model
31 Methodology of Adapting Large English Language Models for Specific Cultural Contexts 提出一种基于指令调优的快速适配方法，用于将大型英语语言模型迁移到特定文化背景。 large language model
32 Poisoned LangChain: Jailbreak LLMs by LangChain 提出 Poisoned-LangChain，通过恶意知识库实现对LLM的间接越狱攻击 large language model
33 Decoding with Limited Teacher Supervision Requires Understanding When to Trust the Teacher 提出自适应信任解码算法，在有限监督下提升小规模LLM生成质量 large language model ✅
34 Implicit Discourse Relation Classification For Nigerian Pidgin 针对尼日利亚皮钦语，提出一种合成语料库的隐式篇章关系分类方法。 large language model
35 Categorical Syllogisms Revisited: A Review of the Logical Reasoning Abilities of LLMs for Analyzing Categorical Syllogism 系统性分析LLM在三段论推理中的逻辑能力，揭示量词理解瓶颈 large language model
36 "Is ChatGPT a Better Explainer than My Professor?": Evaluating the Explanation Capabilities of LLMs in Conversation Compared to a Human Baseline 评估大型语言模型在对话解释能力上与人类专家的差距 large language model
37 Themis: A Reference-free NLG Evaluation Language Model with Flexibility and Interpretability 提出Themis，一种灵活且可解释的无参考NLG评估语言模型，优于GPT-4。 large language model
38 FactFinders at CheckThat! 2024: Refining Check-worthy Statement Detection with LLMs through Data Pruning 通过数据剪枝优化LLM，提升政治文本中待核实陈述的检测性能 large language model
39 Hierarchical Context Pruning: Optimizing Real-World Code Completion with Repository-Level Pretrained Code LLMs 提出层级上下文剪枝（HCP）策略，优化仓库级预训练代码大模型在真实场景下的代码补全。 large language model ✅
40 "Vorbeşti Româneşte?" A Recipe to Train Powerful Romanian LLMs with English Instructions 提出一种基于英语指令微调的罗马尼亚语LLM训练方法，并开源相关资源 large language model
41 Llamipa: An Incremental Discourse Parser Llamipa：提出一种基于LLM微调的增量式篇章分析器，提升下游任务性能。 large language model
42 UIO-LLMs: Unbiased Incremental Optimization for Long-Context LLMs UIO-LLMs：面向长文本LLM的无偏增量优化方法 large language model
43 SafeAligner: Safety Alignment against Jailbreak Attacks via Response Disparity Guidance SafeAligner：通过响应差异引导，增强LLM抵抗越狱攻击的安全性对齐方法 large language model
44 Evaluating Quality of Answers for Retrieval-Augmented Generation: A Strong LLM Is All You Need 提出vRAG-Eval评估框架，利用大语言模型评估RAG应用答案质量 large language model

🔬 支柱七：动作重定向 (Motion Retargeting) (1 篇)

#题目一句话要点标签🔗⭐
45 LOOK-M: Look-Once Optimization in KV Cache for Efficient Multimodal Long-Context Inference 提出LOOK-M，用于高效多模态长文本推理的KV缓存单次优化。 spatial relationship large language model multimodal

🔬 支柱二：RL算法与架构 (RL & Architecture) (1 篇)

#题目一句话要点标签🔗⭐
46 Selective Prompting Tuning for Personalized Conversations with LLMs 提出选择性Prompt调优(SPT)方法，提升LLM在个性化对话中的多样性。 contrastive learning large language model ✅

⬅️ 返回 cs.CL 首页 · 🏠 返回主页