cs.CL（2025-04-15）

📊 共 25 篇论文 | 🔗 7 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (17 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (8 🔗3)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (17 篇)

#题目一句话要点标签🔗⭐
1 Reimagining Urban Science: Scaling Causal Inference with Large Language Models 提出UrbanCIA：利用大语言模型赋能城市科学因果推断，实现可扩展、可复现的城市研究。 large language model multimodal
2 MuSeD: A Multimodal Spanish Dataset for Sexism Detection in Social Media Videos 提出MuSeD多模态西班牙语数据集，用于社交媒体视频中的性别歧视检测。 large language model multimodal
3 Assessment of Evolving Large Language Models in Upper Secondary Mathematics 评估大型语言模型在高中数学考试中的能力演进 large language model
4 Masculine Defaults via Gendered Discourse in Podcasts and Large Language Models 提出性别化语篇相关框架与语篇词嵌入关联测试，揭示播客和LLM中的男性默认偏见。 large language model
5 RankAlign: A Ranking View of the Generator-Validator Gap in Large Language Models RankAlign：通过排序视角解决大语言模型中生成器-验证器之间的差距 large language model
6 Propaganda via AI? A Study on Semantic Backdoors in Large Language Models 提出RAVEN框架，用于检测大型语言模型中基于语义的后门攻击。 large language model ✅
7 Cancer-Myth: Evaluating Large Language Models on Patient Questions with False Presuppositions Cancer-Myth：评估大型语言模型处理含错误预设的患者提问能力 large language model
8 Dependency Structure Augmented Contextual Scoping Framework for Multimodal Aspect-Based Sentiment Analysis 提出DASCO框架，利用依赖结构增强上下文范围，解决多模态情感分析中的关键挑战。 multimodal ✅
9 Benchmarking Next-Generation Reasoning-Focused Large Language Models in Ophthalmology: A Head-to-Head Evaluation on 5,888 Items 眼科领域推理型大语言模型基准测试：5888项的对比评估 large language model
10 A Large-Language Model Framework for Relative Timeline Extraction from PubMed Case Reports 提出基于大语言模型的框架，从PubMed病例报告中提取相对时间线 large language model TAMP
11 Improving Instruct Models for Free: A Study on Partial Adaptation 通过部分适配提升指令模型性能，平衡指令遵循与上下文学习能力。 instruction following
12 TextArena TextArena：用于训练和评估LLM智能行为的竞争性文本游戏平台 large language model ✅
13 Reinforcing Compositional Retrieval: Retrieving Step-by-Step for Composing Informative Contexts 提出基于强化学习的组合式检索框架，用于构建信息丰富的上下文。 large language model
14 The Obvious Invisible Threat: LLM-Powered GUI Agents' Vulnerability to Fine-Print Injections 提出针对LLM驱动的GUI代理的防御策略以解决隐私安全问题 large language model
15 LazyReview A Dataset for Uncovering Lazy Thinking in NLP Peer Reviews 提出LazyReview数据集，用于检测NLP同行评审中的惰性思维。 large language model ✅
16 Understanding LLMs' Cross-Lingual Context Retrieval: How Good It Is And Where It Comes From 评估并解析大语言模型跨语言上下文检索能力及其形成机制 large language model
17 Moving Beyond Next-Token Prediction: Transformers are Context-Sensitive Language Generators 将Transformer解构为上下文敏感语言生成器，突破下一token预测的局限 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (8 篇)

#题目一句话要点标签🔗⭐
18 A Dual-Space Framework for General Knowledge Distillation of Large Language Models 提出双空间知识蒸馏框架DSKD，解决大语言模型通用知识蒸馏问题。 distillation large language model instruction following
19 DeepMLF: Multimodal language model with learnable tokens for deep fusion in sentiment analysis DeepMLF：一种基于可学习Token的多模态语言模型，用于情感分析中的深度融合 representation learning multimodal
20 Dynamic Compressing Prompts for Efficient Inference of Large Language Models 提出动态压缩提示（LLM-DCP）方法，高效推理大型语言模型，显著降低计算成本。 curriculum learning large language model ✅
21 Minitron-SSM: Efficient Hybrid Language Model Compression through Group-Aware SSM Pruning Minitron-SSM：通过分组感知SSM剪枝实现高效混合语言模型压缩 SSM state space model distillation
22 Efficient Reasoning Models: A Survey 综述高效推理模型，加速Chain-of-Thoughts范式在复杂逻辑任务中的应用。 reinforcement learning distillation chain-of-thought ✅
23 ReTool: Reinforcement Learning for Strategic Tool Use in LLMs ReTool：强化学习驱动LLM战略性工具使用，提升复杂数学推理能力 reinforcement learning
24 OpenTuringBench: An Open-Model-based Benchmark and Framework for Machine-Generated Text Detection and Attribution 提出OpenTuringBench，用于评估和训练机器生成文本检测与溯源模型。 contrastive learning large language model ✅
25 ReZero: Enhancing LLM search ability by trying one-more-time ReZero：通过奖励重试机制提升LLM的检索能力 reinforcement learning large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页