cs.CL（2025-04-08）

📊 共 25 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) (24 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (1 🔗1)

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

#题目一句话要点标签🔗⭐
1 From 128K to 4M: Efficient Training of Ultra-Long Context Large Language Models 提出高效训练方法，将LLM上下文长度扩展至4M tokens，并保持性能。 large language model multimodal instruction following ✅
2 Can LLMs Simulate Personas with Reversed Performance? A Benchmark for Counterfactual Instruction Following 提出Counterfactual Instruction Following基准，评估LLM在逆向性能角色模拟中的能力 large language model instruction following
3 Separator Injection Attack: Uncovering Dialogue Biases in Large Language Models Caused by Role Separators 提出分隔符注入攻击(SIA)，揭示角色分隔符导致的大语言模型对话偏见。 large language model instruction following
4 BiasCause: Evaluate Socially Biased Causal Reasoning of Large Language Models BiasCause：评估大型语言模型中社会偏见的因果推理 large language model
5 Exposure to Content Written by Large Language Models Can Reduce Stigma Around Opioid Use Disorder in Online Communities 利用大型语言模型减少在线社区中对鸦片类药物使用障碍的污名化 large language model
6 Assessing how hyperparameters impact Large Language Models' sarcasm detection performance 研究超参数对大型语言模型讽刺检测性能的影响，Llama-2-13b微调后达到人类水平。 large language model
7 Llama-3-Nanda-10B-Chat: An Open Generative Large Language Model for Hindi Nanda：面向印地语的10B参数开源生成式大语言模型，性能领先 large language model
8 Rank-Then-Score: Enhancing Large Language Models for Automated Essay Scoring 提出Rank-Then-Score框架，提升大语言模型在自动作文评分任务上的性能 large language model
9 Don't Let It Hallucinate: Premise Verification via Retrieval-Augmented Logical Reasoning 提出基于检索增强逻辑推理的框架，解决大语言模型中的虚假前提幻觉问题 large language model
10 Language-Dependent Political Bias in AI: A Study of ChatGPT and Gemini 研究揭示ChatGPT和Gemini在不同语言中存在政治倾向性偏差。 large language model
11 S'MoRE: Structural Mixture of Residual Experts for Parameter-Efficient LLM Fine-tuning 提出S'MoRE以解决大语言模型微调的参数效率与模型能力平衡问题 large language model ✅
12 Query Understanding in LLM-based Conversational Information Seeking 探索LLM在对话式信息检索中提升查询理解的技术 large language model
13 Encoder-Decoder Gemma: Improving the Quality-Efficiency Trade-Off via Adaptation 提出Encoder-Decoder Gemma，通过模型适配提升质量-效率权衡。 large language model
14 Navigating the Rabbit Hole: Emergent Biases in LLM-Generated Attack Narratives Targeting Mental Health Groups 揭示LLM针对精神健康群体的攻击性叙事中涌现的偏见 large language model
15 QGen Studio: An Adaptive Question-Answer Generation, Training and Evaluation Platform QGen Studio：一个自适应的问答生成、训练与评估平台 large language model
16 NativQA Framework: Enabling LLMs with Native, Local, and Everyday Knowledge NativQA框架：利用本地化知识赋能大语言模型 large language model ✅
17 Unsupervised Location Mapping for Narrative Corpora 提出一种无监督位置映射方法，用于在叙事语料库中定位故事轨迹。 large language model
18 Enhancing Coreference Resolution with Pretrained Language Models: Bridging the Gap Between Syntax and Semantics 利用预训练语言模型融合句法和语义信息，提升共指消解性能 large language model
19 Leveraging Robust Optimization for LLM Alignment under Distribution Shifts 提出基于鲁棒优化的LLM对齐框架，提升分布偏移下的性能 large language model
20 It's the same but not the same: Do LLMs distinguish Spanish varieties? 评估大型语言模型区分西班牙语变体的能力，发现GPT-4o表现最佳。 large language model
21 SEA-LION: Southeast Asian Languages in One Network 提出SEA-LION，一个面向东南亚语言的先进多语言LLM。 large language model
22 LLM$\times$MapReduce-V2: Entropy-Driven Convolutional Test-Time Scaling for Generating Long-Form Articles from Extremely Long Resources 提出LLM×MapReduce-V2，通过卷积缩放增强LLM处理超长文本生成长文的能力 large language model ✅
23 STRIVE: A Think & Improve Approach with Iterative Refinement for Enhancing Question Quality Estimation 提出STRIVE，通过迭代改进提升大语言模型在问题质量评估中的表现 large language model
24 Towards Smarter Hiring: Are Zero-Shot and Few-Shot Pre-trained LLMs Ready for HR Spoken Interview Transcript Analysis? 评估零样本/少样本LLM在HR面试转录分析中的能力，揭示其局限性 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🔗⭐
25 Multi-Sense Embeddings for Language Models and Knowledge Distillation 提出多义嵌入和知识蒸馏方法，提升语言模型效率并保持性能。 distillation large language model ✅

⬅️ 返回 cs.CL 首页 · 🏠 返回主页