cs.CL（2025-02-18）

📊 共 21 篇论文 | 🔗 1 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (20 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (20 篇)

#题目一句话要点标签🔗⭐
1 Stepwise Perplexity-Guided Refinement for Efficient Chain-of-Thought Reasoning in Large Language Models 提出基于困惑度的逐步优化方法，提升大语言模型CoT推理效率 large language model chain-of-thought
2 Beyond Words: Exploring Cultural Value Sensitivity in Multimodal Models 评估多模态模型中的文化价值观敏感性，揭示其与文化价值对齐的复杂性。 large language model multimodal
3 Facilitating Long Context Understanding via Supervised Chain-of-Thought Reasoning 提出基于监督式思维链推理的长文本理解方法，并构建金融领域合成数据集LongFinanceQA。 large language model chain-of-thought
4 Robust Adaptation of Large Multimodal Models for Retrieval Augmented Hateful Meme Detection 提出一种鲁棒的大型多模态模型自适应框架，用于检索增强的仇恨模因检测。 multimodal ✅
5 When People are Floods: Analyzing Dehumanizing Metaphors in Immigration Discourse with Large Language Models 提出一种结合词级和文档级信号的新方法，利用大型语言模型分析移民讨论中的隐喻 large language model
6 UniGuardian: A Unified Defense for Detecting Prompt Injection, Backdoor Attacks and Adversarial Attacks in Large Language Models UniGuardian：用于检测大语言模型中提示注入、后门攻击和对抗攻击的统一防御机制 large language model
7 STEER-ME: Assessing the Microeconomic Reasoning of Large Language Models STEER-ME：评估大型语言模型在微观经济学推理方面的能力 large language model
8 Towards Text-Image Interleaved Retrieval 提出文本-图像交错检索任务与MME模型，解决多图文场景下的信息检索问题。 large language model multimodal
9 Language Models Can Predict Their Own Behavior 利用语言模型内部表征，无需生成token即可预测其行为，降低风险和加速推理。 instruction following chain-of-thought
10 Grounding LLM Reasoning with Knowledge Graphs 提出基于知识图谱的LLM推理框架，提升推理准确性和可解释性 large language model chain-of-thought
11 Natural Language Generation from Visual Events: State-of-the-Art and Key Open Questions 综述视觉事件到自然语言生成：分析现有方法并探讨关键开放问题 multimodal
12 LLMPopcorn: An Empirical Study of LLMs as Assistants for Popular Micro-video Generation LLMPopcorn：探索大语言模型辅助生成高流量微视频的潜力与方法 large language model
13 Language Models are Few-Shot Graders 提出基于LLM的自动短答案评分（ASAG）流程，提升评分准确性和效率。 large language model
14 Training Turn-by-Turn Verifiers for Dialogue Tutoring Agents: The Curious Case of LLMs as Your Coding Tutors 提出Trace-and-Verify框架，用于训练基于LLM的对话式代码辅导Agent。 large language model
15 Evaluating and Enhancing Out-of-Domain Generalization of Task-Oriented Dialog Systems for Task Completion without Turn-level Dialog Annotations 提出ZeroToD框架，提升零样本任务型对话系统在未见领域的任务完成度。 large language model
16 Improving Multi-turn Task Completion in Task-Oriented Dialog Systems via Prompt Chaining and Fine-Grained Feedback RealTOD框架通过提示链和细粒度反馈，显著提升面向任务对话系统中多轮任务完成的可靠性。 large language model
17 Multilingual Language Model Pretraining using Machine-translated Data 利用机器翻译数据预训练多语言模型，显著提升非英语语言性能。 large language model
18 Neural Attention Search 提出神经注意力搜索(NAtS)框架，用于降低Transformer模型推理时KV缓存大小，从而降低推理成本。 large language model
19 RuozhiBench: Evaluating LLMs with Logical Fallacies and Misleading Premises RuozhiBench：构建逻辑谬误和误导性前提的评测基准，评估LLM的推理能力 large language model
20 Adapting Psycholinguistic Research for LLMs: Gender-inclusive Language in a Coreference Context 研究LLM对性别包容性语言的理解：揭示核心指代中的性别偏见 large language model

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

#题目一句话要点标签🔗⭐
21 Thinking Outside the (Gray) Box: A Context-Based Score for Assessing Value and Originality in Neural Text Generation 提出基于上下文的评分方法，提升神经文本生成中的价值和原创性 reinforcement learning large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页