cs.CL（2025-01-23）

📊 共 39 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) (34 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (3) 支柱一：机器人控制 (Robot Control) (1) 支柱三：空间感知与语义 (Perception & Semantics) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (34 篇)

#题目一句话要点标签🔗⭐
1 Towards Safer Social Media Platforms: Scalable and Performant Few-Shot Harmful Content Moderation Using Large Language Models 利用大语言模型进行少样本有害内容审核，提升社交媒体平台安全性 large language model multimodal
2 Re-ranking Using Large Language Models for Mitigating Exposure to Harmful Content on Social Media Platforms 提出基于大语言模型的重排序方法，缓解社交媒体平台有害内容暴露问题 large language model
3 RAMQA: A Unified Framework for Retrieval-Augmented Multi-Modal Question Answering 提出RAMQA，一个统一的检索增强多模态问答框架，提升生成式LLM在MRAQA任务中的性能。 large language model multimodal ✅
4 Parameter-Efficient Fine-Tuning for Foundation Models 提出参数高效微调方法以优化基础模型性能 foundation model multimodal ✅
5 Do Large Language Models Truly Understand Geometric Structures? 提出GeomRel数据集与GeoCoT方法，提升大语言模型对几何结构的理解能力 large language model chain-of-thought
6 MedSlice: Fine-Tuned Large Language Models for Secure Clinical Note Sectioning 提出MedSlice以解决临床笔记分段的隐私与效率问题 large language model
7 Predicting Compact Phrasal Rewrites with Large Language Models for ASR Post Editing 提出基于短语的压缩重写方法，提升LLM在ASR后编辑任务中的效率与精度 large language model
8 Softplus Attention with Re-weighting Boosts Length Extrapolation in Large Language Models 提出Softplus注意力机制与重加权策略，显著提升大语言模型长度外推能力 large language model
9 UGMathBench: A Diverse and Dynamic Benchmark for Undergraduate-Level Mathematical Reasoning with Large Language Models UGMathBench：一个用于评估大语言模型本科数学推理能力的多元动态基准 large language model ✅
10 Musical ethnocentrism in Large Language Models 揭示大型语言模型中的音乐民族中心主义偏见 large language model
11 Framework for Progressive Knowledge Fusion in Large Language Models Through Structured Conceptual Redundancy Analysis 提出一种基于结构化概念冗余分析的大语言模型渐进式知识融合框架 large language model
12 CAPRAG: A Large Language Model Solution for Customer Service and Automatic Reporting using Vector and Graph Retrieval-Augmented Generation CAPRAG：利用向量和图检索增强生成技术，为银行客户服务和自动报告提供大语言模型解决方案 large language model
13 AdEval: Alignment-based Dynamic Evaluation to Mitigate Data Contamination in Large Language Models AdEval：一种基于对齐的动态评估方法，用于缓解大语言模型中的数据污染问题 large language model
14 Emotions, Context, and Substance Use in Adolescents: A Large Language Model Analysis of Reddit Posts 利用大型语言模型分析Reddit帖子，揭示青少年情绪、环境与物质使用间的关联。 large language model
15 DI-BENCH: Benchmarking Large Language Models on Dependency Inference with Testable Repositories at Scale DI-BENCH：大规模依赖推断基准测试，评估大语言模型在软件仓库上的性能 large language model
16 LVPruning: An Effective yet Simple Language-Guided Vision Token Pruning Approach for Multi-modal Large Language Models 提出LVPruning，通过语言引导的视觉Token剪枝，高效压缩多模态大语言模型。 large language model
17 Improving Contextual Faithfulness of Large Language Models via Retrieval Heads-Induced Optimization 提出RHIO框架，通过检索头优化提升大语言模型在长文本问答中的上下文忠实性 large language model
18 Can Large Language Models Understand Preferences in Personalized Recommendation? 提出PerRecBench，评估LLM在消除用户评分偏差和物品质量影响下的个性化推荐能力。 large language model ✅
19 Do as We Do, Not as You Think: the Conformity of Large Language Models 提出BenchForm基准，研究LLM多智能体系统中从众行为，并探索缓解策略。 large language model
20 Advancing Mathematical Reasoning in Language Models: The Impact of Problem-Solving Data, Data Synthesis Methods, and Training Stages 通过问题求解数据、数据合成方法和训练阶段优化语言模型中的数学推理能力 large language model instruction following
21 Explainable XR: Understanding User Behaviors of XR Environments using LLM-assisted Analytics Framework 提出Explainable XR框架，利用LLM辅助分析XR环境中的用户行为 large language model multimodal
22 OSUM: Advancing Open Speech Understanding Models with Limited Resources in Academia 提出OSUM以解决学术界资源有限下的语音理解问题 large language model TAMP
23 Pseudocode-Injection Magic: Enabling LLMs to Tackle Graph Computational Tasks PIE：通过伪代码注入增强LLM在图计算任务中的推理能力 large language model
24 LLMs Can Plan Only If We Tell Them 提出AoT+算法，提升LLM在复杂规划任务中的自主规划能力，超越人类基线。 large language model
25 RECALL: Library-Like Behavior In Language Models is Enhanced by Self-Referencing Causal Cycles 提出RECALL机制，通过自引用因果循环增强语言模型的记忆能力。 large language model
26 Chain of Grounded Objectives: Bridging Process and Goal-oriented Prompting for Code Generation 提出Chain of Grounded Objectives (CGO)，提升LLM在代码生成任务中的性能 large language model
27 Watching the AI Watchdogs: A Fairness and Robustness Analysis of AI Safety Moderation Classifiers 分析AI安全审核分类器的公平性和鲁棒性，揭示潜在差距 large language model
28 LLMs are Vulnerable to Malicious Prompts Disguised as Scientific Language LLM易受伪装成科学语言的恶意提示攻击，导致偏见和虚假信息生成 large language model
29 The Breeze 2 Herd of Models: Traditional Chinese LLMs Based on Llama with Vision-Aware and Function-Calling Capabilities 联发科发布Breeze2模型系列：支持繁体中文、视觉理解和函数调用的Llama模型 instruction following
30 Analysis of Indic Language Capabilities in LLMs 评估LLM在印度语言上的能力，为安全基准测试选择合适语言。 large language model
31 QuanTaxo: A Quantum Approach to Self-Supervised Taxonomy Expansion 提出QuanTaxo，一种量子启发的自监督分类扩展方法，提升层级多义性建模能力。 PaLM-E
32 A RAG-Based Institutional Assistant 构建基于RAG的机构助手，提升大型语言模型在知识密集型任务中的表现 large language model
33 Sigma: Differential Rescaling of Query, Key and Value for Efficient Language Models Sigma：通过差异化重缩放QKV提升语言模型效率，专为系统领域设计。 large language model
34 Hypothesis Generation for Materials Discovery and Design Using Goal-Driven and Constraint-Guided LLM Agents 提出基于目标驱动和约束引导的LLM代理以加速材料发现 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (3 篇)

#题目一句话要点标签🔗⭐
35 CRPO: Confidence-Reward Driven Preference Optimization for Machine Translation 提出CRPO方法，结合置信度和奖励优化机器翻译偏好学习，提升数据效率和翻译精度。 reinforcement learning RLHF DPO
36 ExLM: Rethinking the Impact of [MASK] Tokens in Masked Language Models ExLM：通过增强上下文缓解掩码语言模型中[MASK]引入的语义歧义问题 representation learning multimodal
37 Multi-Level Attention and Contrastive Learning for Enhanced Text Classification with an Optimized Transformer 提出多层注意力与对比学习优化的Transformer，提升文本分类性能 contrastive learning

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
38 Autonomous Structural Memory Manipulation for Large Language Models Using Hierarchical Embedding Augmentation 提出基于层级嵌入增强的自主结构记忆操纵方法，提升大语言模型在复杂任务中的效率与泛化能力。 manipulation large language model

🔬 支柱三：空间感知与语义 (Perception & Semantics) (1 篇)

#题目一句话要点标签🔗⭐
39 RPO: Retrieval Preference Optimization for Robust Retrieval-Augmented Generation 提出检索偏好优化(RPO)方法，提升检索增强生成模型在知识冲突下的鲁棒性。 implicit representation large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页