cs.CL（2025-08-26）

📊 共 28 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (23 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (4) 支柱六：视频提取与匹配 (Video Extraction) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (23 篇)

#题目一句话要点标签🔗⭐
1 Tailored Teaching with Balanced Difficulty: Elevating Reasoning in Multimodal Chain-of-Thought via Prompt Curriculum 提出基于平衡难度的定制教学方法以提升多模态推理能力 large language model multimodal chain-of-thought
2 Empathy Omni: Enabling Empathetic Speech Response Generation through Large Language Models 提出Emotion Omni以解决情感理解不足的问题 large language model instruction following ✅
3 Scaling Laws for Task-Stratified Knowledge in Post-Training Quantized Large Language Models 提出任务分层知识的缩放规律以优化后训练量化大语言模型 large language model
4 Automatic Question & Answer Generation Using Generative Large Language Model (LLM) 提出基于生成大语言模型的自动化问答生成方法以简化教育评估 large language model
5 LongReasonArena: A Long Reasoning Benchmark for Large Language Models 提出LongReasonArena以评估大型语言模型的长推理能力 large language model ✅
6 ConfTuner: Training Large Language Models to Express Their Confidence Verbally 提出ConfTuner以解决大语言模型信心表达不准确问题 large language model ✅
7 M3HG: Multimodal, Multi-scale, and Multi-type Node Heterogeneous Graph for Emotion Cause Triplet Extraction in Conversations 提出M3HG以解决多模态对话中的情感原因三元组提取问题 multimodal ✅
8 Breaking the Trade-Off Between Faithfulness and Expressiveness for Large Language Models 提出协作解码方法以解决大语言模型的忠实性与表现力权衡问题 large language model
9 Arrows of Math Reasoning Data Synthesis for Large Language Models: Diversity, Complexity and Correctness 提出程序辅助合成框架以提升大语言模型的数学推理能力 large language model
10 Topic Identification in LLM Input-Output Pairs through the Lens of Information Bottleneck 提出UDIB方法以提高LLM输入输出对的主题识别能力 large language model
11 ReflectivePrompt: Reflective evolution in autoprompting algorithms 提出ReflectivePrompt以优化语言模型的自动提示选择 large language model
12 A perishable ability? The future of writing in the face of generative artificial intelligence 探讨生成性人工智能对人类写作能力的影响 large language model
13 Reflective Agreement: Combining Self-Mixture of Agents with a Sequence Tagger for Robust Event Extraction 提出ARIS以解决事件提取中的精度与召回率问题 large language model
14 Generative Interfaces for Language Models 提出生成接口以解决语言模型交互效率低下问题 large language model
15 HiPlan: Hierarchical Planning for LLM-Based Agents with Adaptive Global-Local Guidance 提出HiPlan以解决LLM代理在复杂规划中的决策问题 large language model
16 Diverse And Private Synthetic Datasets Generation for RAG evaluation: A multi-agent framework 提出多智能体框架以生成多样化且保护隐私的合成数据集用于RAG评估 large language model
17 What do language models model? Transformers, automata, and the format of thought 探讨语言模型的本质及其与人类认知的关系 large language model
18 Beyond the Black Box: Integrating Lexical and Semantic Methods in Quantitative Discourse Analysis with BERTopic 提出透明框架以提升定量话语分析的有效性与可重复性 large language model
19 It's All About In-Context Learning! Teaching Extremely Low-Resource Languages to LLMs 提出基于上下文学习的方法以解决极低资源语言的支持问题 large language model
20 The Double-edged Sword of LLM-based Data Reconstruction: Understanding and Mitigating Contextual Vulnerability in Word-level Differential Privacy Text Sanitization 提出LLM辅助的数据重构方法以缓解文本隐私问题 large language model
21 Empowering Computing Education Researchers Through LLM-Assisted Content Analysis 提出LLM辅助内容分析方法以解决计算教育研究中的数据分析挑战 large language model
22 Controllable Conversational Theme Detection Track at DSTC 12 提出可控对话主题检测以解决对话分析中的主题识别问题 large language model
23 Thinking Before You Speak: A Proactive Test-time Scaling Approach 提出TBYS框架以解决复杂推理任务中的思维缺失问题 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (4 篇)

#题目一句话要点标签🔗⭐
24 ThinkDial: An Open Recipe for Controlling Reasoning Effort in Large Language Models 提出ThinkDial以解决大语言模型推理控制问题 reinforcement learning reward shaping large language model
25 Beyond Quality: Unlocking Diversity in Ad Headline Generation with Large Language Models 提出DIVER框架以解决广告标题生成中的多样性问题 reinforcement learning large language model
26 Automatic Prompt Optimization with Prompt Distillation 提出DistillPrompt以优化语言模型的自动提示选择 distillation large language model
27 Harnessing Rule-Based Reinforcement Learning for Enhanced Grammatical Error Correction 提出基于规则的强化学习框架以提升语法错误纠正效果 reinforcement learning

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

#题目一句话要点标签🔗⭐
28 One Joke to Rule them All? On the (Im)possibility of Generalizing Humor 提出幽默类型迁移学习方法以解决幽默理解问题 HuMoR large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页