cs.CL（2025-07-15）

📊 共 24 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) (24 🔗4)

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

#题目一句话要点标签🔗⭐
1 MapIQ: Evaluating Multimodal Large Language Models for Map Question Answering 提出MapIQ基准数据集，评估多模态大语言模型在地图问答中的能力 large language model multimodal
2 MetaLint: Generalizable Idiomatic Code Quality Analysis through Instruction-Following and Easy-to-Hard Generalization MetaLint：通过指令跟随和由易到难泛化实现通用代码质量分析 large language model instruction following
3 MSA at ImageCLEF 2025 Multimodal Reasoning: Multilingual Multimodal Reasoning With Ensemble Vision Language Models 提出基于集成视觉语言模型的MSA多语言多模态推理系统，在ImageCLEF 2025挑战赛中取得领先。 large language model multimodal
4 Reasoning Strategies in Large Language Models: Can They Follow, Prefer, and Optimize? 研究大型语言模型推理策略控制与优化，提升逻辑问题解决能力 large language model
5 EXAONE 4.0: Unified Large Language Models Integrating Non-reasoning and Reasoning Modes EXAONE 4.0：融合非推理与推理模式的统一大型语言模型 large language model ✅
6 What is the Best Process Model Representation? A Comparative Analysis for Process Modeling with Large Language Models 对比分析不同流程模型表示，为大语言模型流程建模任务选择最优方案 large language model
7 Automated Novelty Evaluation of Academic Paper: A Collaborative Approach Integrating Human and Large Language Model Knowledge 提出融合人类专家与大语言模型知识的学术论文新颖性自动评估方法 large language model
8 Internal Value Alignment in Large Language Models through Controlled Value Vector Activation 提出ConVA方法，通过控制价值向量激活实现大语言模型内部价值对齐 large language model ✅
9 LRCTI: A Large Language Model-Based Framework for Multi-Step Evidence Retrieval and Reasoning in Cyber Threat Intelligence Credibility Verification LRCTI：基于大语言模型的多步证据检索与推理框架，用于网络威胁情报可信度验证 large language model
10 Persona-Based Synthetic Data Generation Using Multi-Stage Conditioning with Large Language Models for Emotion Recognition PersonaGen：基于多阶段条件LLM的合成数据生成，用于情感识别 large language model
11 Evaluating Speech-to-Text x LLM x Text-to-Speech Combinations for AI Interview Systems 评估STT x LLM x TTS组合在AI面试系统中的性能 large language model multimodal
12 ExpliCIT-QA: Explainable Code-Based Image Table Question Answering 提出ExplicIT-QA，解决图像表格问答的可解释性问题。 multimodal chain-of-thought
13 KisMATH: Do LLMs Have Knowledge of Implicit Structures in Mathematical Reasoning? KisMATH：探究LLM在数学推理中对隐式结构的认知能力，并提出因果CoT图进行分析。 large language model chain-of-thought
14 An Agentic Flow for Finite State Machine Extraction using Prompt Chaining FlowFSM：利用提示链的大语言模型自动提取有限状态机 large language model chain-of-thought
15 Partitioner Guided Modal Learning Framework 提出分区引导的模态学习框架以提升多模态学习效果 multimodal
16 SAFT: Structure-Aware Fine-Tuning of LLMs for AMR-to-Text Generation SAFT：一种结构感知的LLM微调方法，用于AMR到文本的生成 large language model
17 Seq vs Seq: An Open Suite of Paired Encoders and Decoders 提出Ettin模型套件，系统性对比Encoder和Decoder架构在不同任务上的性能差异。 large language model
18 FMC: Formalization of Natural Language Mathematical Competition Problems 提出基于大语言模型和误差反馈的自动形式化方法，构建奥林匹克级数学题形式化数据集。 large language model
19 KV-Latent: Dimensional-level KV Cache Reduction with Frequency-aware Rotary Positional Embedding 提出KV-Latent，通过维度降采样和频率感知旋转位置编码，有效降低LLM的KV缓存占用。 large language model ✅
20 Sparse Autoencoders Can Capture Language-Specific Concepts Across Diverse Languages 提出SAE-LAPE方法，利用稀疏自编码器识别LLM中语言特定的概念 large language model ✅
21 Temperature and Persona Shape LLM Agent Consensus With Minimal Accuracy Gains in Qualitative Coding 研究表明，在定性编码中，LLM Agent的温度和角色设定对共识影响显著，但准确率提升有限。 large language model
22 What Should LLMs Forget? Quantifying Personal Data in LLMs for Right-to-Be-Forgotten Requests 提出WikiMem数据集和模型无关指标，量化LLM中个人数据，支持被遗忘权请求 large language model
23 Multi-Trigger Poisoning Amplifies Backdoor Vulnerabilities in LLMs 多触发器投毒攻击放大LLM后门漏洞，提出选择性重训练防御方法 large language model
24 Teach Me Sign: Stepwise Prompting LLM for Sign Language Production 提出TEAM-Sign：利用逐步提示的大语言模型进行手语生成 large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页