cs.CL（2024-11-08）

📊 共 27 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (25 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (2)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (25 篇)

#题目一句话要点标签🔗⭐
1 The Dark Patterns of Personalized Persuasion in Large Language Models: Exposing Persuasive Linguistic Features for Big Five Personality Traits in LLMs Responses 揭示大型语言模型中基于人格特质的个性化说服的“黑暗模式” large language model
2 Exploring the Limits of Large Language Models: A Systematic Evaluation of Masked Text Processing Ability through MskQA and MskCal 通过MskQA和MskCal评估LLM在掩码文本处理中的局限性 large language model
3 Humans and Large Language Models in Clinical Decision Support: A Study with Medical Calculators 评估大型语言模型在临床决策支持中的应用：以医学计算器选择为例 large language model
4 Evaluating Large Language Model Capability in Vietnamese Fact-Checking Data Generation 评估大型语言模型在越南语事实核查数据生成中的能力 large language model
5 Assessing Open-Source Large Language Models on Argumentation Mining Subtasks 评估开源大语言模型在论证挖掘子任务上的能力 large language model
6 Evaluating and Adapting Large Language Models to Represent Folktales in Low-Resource Languages 评估并调整大型语言模型以表示低资源语言中的民间故事 large language model
7 Identifying and Decomposing Compound Ingredients in Meal Plans Using Large Language Models 利用大型语言模型识别和分解膳食计划中的复合成分 large language model
8 LBPE: Long-token-first Tokenization to Improve Large Language Models 提出LBPE：一种长token优先的分词方法，以改善大型语言模型性能。 large language model
9 Benchmarking Distributional Alignment of Large Language Models 构建基准测试，评估大型语言模型在模拟特定人群观点分布上的对齐能力 large language model
10 Gap-Filling Prompting Enhances Code-Assisted Mathematical Reasoning 提出Gap-Filling Prompting，提升小模型在数学推理中的代码辅助能力 large language model chain-of-thought
11 Dynamic-SUPERB Phase-2: A Collaboratively Expanding Benchmark for Measuring the Capabilities of Spoken Language Models with 180 Tasks 发布Dynamic-SUPERB Phase-2：一个协同扩展的语音语言模型能力评估基准，包含180个任务 foundation model multimodal ✅
12 Reasoning Robustness of LLMs to Adversarial Typographical Errors 提出对抗性拼写错误攻击方法，评估LLM推理的鲁棒性 large language model chain-of-thought
13 FactLens: Benchmarking Fine-Grained Fact Verification FactLens：提出细粒度事实核查基准，解决LLM幻觉问题中传统方法的不足。 large language model
14 BERTrend: Neural Topic Modeling for Emerging Trends Detection BERTrend：用于新兴趋势检测的神经主题模型 large language model
15 RefreshKV: Updating Small KV Cache During Long-form Generation RefreshKV：通过动态更新小KV缓存提升长文本生成性能 large language model
16 Fact or Fiction? Can LLMs be Reliable Annotators for Political Truths? 利用大语言模型作为政治真相的可靠标注器，解决政治虚假信息检测问题。 large language model
17 Multi-hop Evidence Pursuit Meets the Web: Team Papelo at FEVER 2024 Papelo团队提出多跳证据追踪方法，结合LLM推理和搜索引擎检索，提升FEVER 2024任务的声明验证效果。 large language model
18 One Small and One Large for Document-level Event Argument Extraction 针对文档级事件论元抽取，提出基于小型和大型语言模型的双重优化方法。 large language model ✅
19 SSSD: Simply-Scalable Speculative Decoding 提出SSSD：一种简单可扩展的无训练推测解码方法，加速大语言模型推理。 large language model
20 Assessing the Answerability of Queries in Retrieval-Augmented Code Generation 提出RaCGEval评估基准，用于评估检索增强代码生成中查询的可回答性。 large language model
21 An Early FIRST Reproduction and Improvements to Single-Token Decoding for Fast Listwise Reranking FIRST：单Token解码加速Listwise重排序，并验证其有效性和效率。 large language model
22 KyrgyzNLP: Challenges, Progress, and Future 关注吉尔吉斯语NLP：挑战、进展与未来展望 large language model
23 EUREKHA: Enhancing User Representation for Key Hackers Identification in Underground Forums EUREKHA：通过增强用户表示，识别地下论坛中的关键黑客。 large language model
24 VISTA: Visual Integrated System for Tailored Automation in Math Problem Generation Using LLM VISTA：利用LLM为数学问题生成定制化自动化视觉集成系统 large language model
25 Towards Low-Resource Harmful Meme Detection with LMM Agents 提出基于LMM Agent的框架，解决低资源有害Meme检测问题 multimodal

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

#题目一句话要点标签🔗⭐
26 Improving Multi-Domain Task-Oriented Dialogue System with Offline Reinforcement Learning 提出一种基于离线强化学习的多领域任务型对话系统，提升任务完成率。 reinforcement learning offline reinforcement learning large language model
27 Reducing Distraction in Long-Context Language Models by Focused Learning 提出基于聚焦学习的长文本语言模型优化方法，减少无关信息干扰。 contrastive learning large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页