cs.CL（2024-06-05）

📊 共 32 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) (24 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (6) 支柱一：机器人控制 (Robot Control) (2)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (24 篇)

#题目一句话要点标签🔗⭐
1 NUMCoT: Numerals and Units of Measurement in Chain-of-Thought Reasoning using Large Language Models NUMCoT：利用大语言模型进行链式推理中数字和计量单位的处理研究 large language model chain-of-thought
2 Wings: Learning Multimodal LLMs without Text-only Forgetting Wings：一种解决多模态LLM中文本遗忘问题的新型架构 large language model multimodal
3 Large Language Models as Evaluators for Recommendation Explanations 探索大语言模型作为推荐解释评估器的可行性，提升评估效率与一致性 large language model instruction following ✅
4 Docs2KG: Unified Knowledge Graph Construction from Heterogeneous Documents Assisted by Large Language Models 提出Docs2KG框架，利用大语言模型从异构文档中构建统一知识图谱 large language model multimodal
5 Exploring Multilingual Large Language Models for Enhanced TNM classification of Radiology Report in lung cancer staging 利用多语言大语言模型提升肺癌分期中放射报告的TNM分类 large language model
6 IrokoBench: A New Benchmark for African Languages in the Age of Large Language Models IrokoBench：面向非洲语言的大语言模型评测基准 large language model
7 Queue management for slo-oriented large language model serving QLM：面向松弛延迟需求的LLM服务队列管理系统，提升资源利用率。 large language model
8 Automating Turkish Educational Quiz Generation Using Large Language Models 提出 Turkish-Quiz-Instruct 数据集，并利用大语言模型自动生成土耳其语教育测验 large language model
9 Assessing the Emergent Symbolic Reasoning Abilities of Llama Large Language Models 评估Llama大型语言模型在符号推理方面的涌现能力 large language model
10 FragRel: Exploiting Fragment-level Relations in the External Memory of Large Language Models FragRel：利用片段级关系增强大语言模型外部记忆，提升长文本处理能力 large language model
11 Bi-Chainer: Automated Large Language Models Reasoning with Bidirectional Chaining 提出Bi-Chainer双向链式推理，提升LLM在复杂逻辑问题上的推理精度与效率 large language model
12 Unveiling Selection Biases: Exploring Order and Token Sensitivity in Large Language Models 揭示大语言模型中的选择偏差：探究顺序和Token敏感性 large language model
13 Evaluating the Efficacy of Large Language Models in Detecting Fake News: A Comparative Analysis 对比评估大型语言模型在假新闻检测中的有效性，为信息完整性提供参考。 large language model
14 Improve Mathematical Reasoning in Language Models by Automated Process Supervision 提出OmegaPRM算法，实现数学推理语言模型的自动化过程监督 large language model chain-of-thought
15 Analyzing LLM Behavior in Dialogue Summarization: Unveiling Circumstantial Hallucination Trends 分析LLM在对话摘要中的行为，揭示情境性幻觉趋势 large language model
16 PatentEval: Understanding Errors in Patent Generation PatentEval：提出专利生成错误类型学，用于评估语言模型在专利文本生成任务中的表现 large language model
17 TACT: Advancing Complex Aggregative Reasoning with Information Extraction Tools TACT：利用信息抽取工具提升复杂聚合推理能力 large language model
18 Cycles of Thought: Measuring LLM Confidence through Stable Explanations 提出基于解释稳定性的LLM置信度评估框架，提升不确定性量化效果 large language model
19 StatBot.Swiss: Bilingual Open Data Exploration in Natural Language 发布StatBot.Swiss双语数据集，评估LLM在Text-to-SQL任务中的泛化能力。 large language model
20 Towards Detecting LLMs Hallucination via Markov Chain-based Multi-agent Debate Framework 提出基于马尔可夫链的多智能体辩论框架，用于检测LLM的幻觉问题 large language model
21 Exploring Human-AI Perception Alignment in Sensory Experiences: Do LLMs Understand Textile Hand? 探索人机感知对齐：大型语言模型能否理解纺织品手感？ large language model
22 BadAgent: Inserting and Activating Backdoor Attacks in LLM Agents BadAgent：在LLM Agent中植入和激活后门攻击 large language model ✅
23 HYDRA: Model Factorization Framework for Black-Box LLM Personalization HYDRA：一种用于黑盒LLM个性化的模型分解框架 large language model ✅
24 LLM as a Scorer: The Impact of Output Order on Dialogue Evaluation 研究LLM作为评分器时，输出顺序对对话评估的影响 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🔗⭐
25 Adversarial Moment-Matching Distillation of Large Language Models 提出对抗矩匹配蒸馏方法，提升大语言模型知识蒸馏效果并达到SOTA imitation learning distillation large language model
26 PLaD: Preference-based Large Language Model Distillation with Pseudo-Preference Pairs 提出PLaD框架，利用伪偏好对进行大语言模型蒸馏，解决资源受限场景下的模型部署问题。 teacher-student distillation large language model
27 Knowledge-Infused Legal Wisdom: Navigating LLM Consultation through the Lens of Diagnostics and Positive-Unlabeled Reinforcement Learning 提出D3LM，利用诊断式提问和PURL解决LLM在法律咨询中信息不足问题 reinforcement learning large language model
28 LLM-based Rewriting of Inappropriate Argumentation using Reinforcement Learning from Machine Feedback 提出基于强化学习的LLM重写方法，以减轻在线讨论中不当言论。 reinforcement learning large language model
29 From Tarzan to Tolkien: Controlling the Language Proficiency Level of LLMs for Content Generation 提出CALM模型，通过微调和强化学习对齐，控制LLM生成文本的语言熟练度 reinforcement learning large language model
30 Open Grounded Planning: Challenges and Benchmark Construction 提出开放环境下的具身规划任务与基准，评估LLM在可执行规划中的能力 reinforcement learning large language model

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
31 BIPED: Pedagogically Informed Tutoring System for ESL Education BIPED：一个基于教学法的ESL教育对话式智能辅导系统 biped large language model
32 Ranking Manipulation for Conversational Search Engines 提出针对会话式搜索引擎的排名操纵攻击方法，提升低排名产品 manipulation large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页