cs.CL（2025-05-31）

📊 共 30 篇论文 | 🔗 5 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (28 🔗5) 支柱六：视频提取与匹配 (Video Extraction) (1) 支柱二：RL算法与架构 (RL & Architecture) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (28 篇)

#题目一句话要点标签🔗⭐
1 Con Instruction: Universal Jailbreaking of Multimodal Large Language Models via Non-Textual Modalities Con Instruction：通过非文本模态实现多模态大语言模型的通用越狱 large language model multimodal
2 Enhancing Multimodal Continual Instruction Tuning with BranchLoRA 提出BranchLoRA，增强多模态持续指令调优，缓解灾难性遗忘。 large language model multimodal
3 Chain-of-Thought Training for Open E2E Spoken Dialogue Systems 提出基于思维链（CoT）训练的端到端口语对话系统，提升语义连贯性。 multimodal chain-of-thought
4 Synergizing LLMs with Global Label Propagation for Multimodal Fake News Detection 提出GLPN-LLM，结合LLM伪标签与全局标签传播，提升多模态假新闻检测性能。 large language model multimodal
5 Enabling Chatbots with Eyes and Ears: An Immersive Multimodal Conversation System for Dynamic Interactions 提出基于多模态记忆检索的对话系统，增强chatbot在动态交互中的视听能力 multimodal
6 Translate With Care: Addressing Gender Bias, Neutrality, and Reasoning in Large Language Model Translations 提出Translate-with-Care数据集以解决机器翻译中的性别偏见问题 large language model
7 A Large Language Model Based Pipeline for Review of Systems Entity Recognition from Clinical Notes 提出基于大语言模型的临床笔记ROS实体识别流水线，提升医疗文档处理效率。 large language model
8 Structured Gradient Guidance for Few-Shot Adaptation in Large Language Models 提出结构化梯度引导方法，提升大语言模型在少样本学习中的适应性和稳定性。 large language model
9 Massively Multilingual Adaptation of Large Language Models Using Bilingual Translation Data EMMA-500：利用双语翻译数据实现LLaMA3模型的大规模多语言适配 large language model
10 G2S: A General-to-Specific Learning Framework for Temporal Knowledge Graph Forecasting with Large Language Models 提出G2S框架，解耦通用知识与场景信息，提升LLM在时序知识图谱预测中的泛化能力 large language model
11 Beyond Context to Cognitive Appraisal: Emotion Reasoning as a Theory of Mind Benchmark for Large Language Models 提出基于认知评估理论的心智理论基准，评估LLM的情感推理能力 large language model
12 Auto-Patching: Enhancing Multi-Hop Reasoning in Language Models 提出Auto-Patch以增强语言模型的多跳推理能力 large language model chain-of-thought
13 Measuring Faithfulness and Abstention: An Automated Pipeline for Evaluating LLM-Generated 3-ply Case-Based Legal Arguments 提出自动化评估流程，用于评估LLM生成3层案例法律论证的忠实性和克制能力。 large language model ✅
14 Accelerating Diffusion LLMs via Adaptive Parallel Decoding 提出自适应并行解码(APD)加速扩散LLM，在吞吐量和质量间灵活权衡。 large language model
15 Assortment of Attention Heads: Accelerating Federated PEFT with Head Pruning and Strategic Client Selection 提出基于注意力头剪枝与策略性客户端选择的联邦PEFT加速方法 large language model
16 Decoupling Reasoning and Knowledge Injection for In-Context Knowledge Editing DecKER：解耦推理与知识注入，提升上下文知识编辑性能 large language model ✅
17 CausalAbstain: Enhancing Multilingual LLMs with Causal Reasoning for Trustworthy Abstention CausalAbstain：利用因果推理增强多语言LLM的可信拒答能力 large language model ✅
18 Dual Debiasing for Noisy In-Context Learning for Text Generation 提出双重去偏框架，解决文本生成中噪声上下文学习的偏差问题 large language model
19 How Significant Are the Real Performance Gains? An Unbiased Evaluation Framework for GraphRAG 提出GraphRAG的无偏评估框架，解决现有评估体系的偏差问题，更准确评估性能。 large language model
20 Efficient Latent Semantic Clustering for Scaling Test-Time Computation of LLMs 提出LSC：利用LLM内部隐状态进行高效语义聚类，加速LLM测试时计算。 large language model
21 DefenderBench: A Toolkit for Evaluating Language Agents in Cybersecurity Environments DefenderBench：用于评估语言智能体在网络安全环境中表现的工具包 large language model ✅
22 GuideX: Guided Synthetic Data Generation for Zero-Shot Information Extraction GuideX：引导式合成数据生成，用于零样本信息抽取 large language model
23 Social Construction of Urban Space: Using LLMs to Identify Neighborhood Boundaries From Craigslist Ads 利用大型语言模型从Craigslist广告中识别城市社区边界，揭示城市空间的社会构建。 large language model
24 Smotrom tvoja pa ander drogoj verden! Resurrecting Dead Pidgin with Generative Models: Russenorsk Case Study 利用生成模型重构已消亡的Russenorsk语：案例研究 large language model
25 Evaluating the Evaluation of Diversity in Commonsense Generation 针对常识生成多样性评估，提出基于内容的评估指标优于形式评估指标的结论。 large language model
26 Goal-Aware Identification and Rectification of Misinformation in Multi-Agent Systems 提出ARGUS框架，用于多智能体系统中目标导向的错误信息识别与修正 large language model ✅
27 Speculative Reward Model Boosts Decision Making Ability of LLMs Cost-Effectively 提出推测奖励模型(SRM)，在降低计算成本的同时提升LLM的决策能力 large language model
28 OWSM v4: Improving Open Whisper-Style Speech Models via Data Scaling and Cleaning OWSM v4：通过数据规模扩展和清洗提升Open Whisper-Style语音模型 foundation model

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

#题目一句话要点标签🔗⭐
29 Sarc7: Evaluating Sarcasm Detection and Generation with Seven Types and Emotion-Informed Techniques Sarc7：提出七种讽刺类型基准，并用情感信息技术改进讽刺检测与生成 HuMoR large language model chain-of-thought

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

#题目一句话要点标签🔗⭐
30 ARIA: Training Language Agents with Intention-Driven Reward Aggregation ARIA：通过意图驱动的奖励聚合训练语言Agent reinforcement learning large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页