cs.CL（2023-10-02）

📊 共 29 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 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (5 🔗2)

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

#题目一句话要点标签🔗⭐
1 Back to the Future: Towards Explainable Temporal Reasoning with Large Language Models 提出可解释的时间推理任务以解决复杂事件预测问题 large language model instruction following
2 Application of frozen large-scale models to multimodal task-oriented dialogue 利用LENS框架提升多模态任务导向对话的性能 large language model multimodal
3 LoFT: Local Proxy Fine-tuning For Improving Transferability Of Adversarial Attacks Against Large Language Model 提出LoFT以提高对大型语言模型的对抗攻击可转移性 large language model
4 Synthetic Data Generation in Low-Resource Settings via Fine-Tuning of Large Language Models 通过微调大语言模型生成合成数据以解决低资源问题 large language model
5 Necessary and Sufficient Watermark for Large Language Models 提出必要且充分的水印技术以解决大语言模型文本识别问题 large language model
6 FedBPT: Efficient Federated Black-box Prompt Tuning for Large Language Models 提出FedBPT以解决联邦学习中大语言模型微调的隐私与效率问题 large language model
7 Knowledge Crosswords: Geometric Knowledge Reasoning with Large Language Models 提出知识填字游戏以解决几何知识推理问题 large language model
8 Reasoning on Graphs: Faithful and Interpretable Large Language Model Reasoning 提出图推理方法以解决大型语言模型的可信性与可解释性问题 large language model
9 All Languages Matter: On the Multilingual Safety of Large Language Models 提出XSafety基准以解决多语言大语言模型安全性问题 large language model ✅
10 Resolving Knowledge Conflicts in Large Language Models 提出评估框架以解决大型语言模型中的知识冲突问题 large language model
11 TRAM: Benchmarking Temporal Reasoning for Large Language Models 提出TRAM基准以评估大语言模型的时间推理能力 large language model
12 ARN: Analogical Reasoning on Narratives 提出ARN框架以解决叙事中的类比推理问题 large language model chain-of-thought
13 PACIT: Unlocking the Power of Examples for Better In-Context Instruction Tuning 提出PACIT以提升上下文指令调优效果 large language model instruction following
14 GraphText: Graph Reasoning in Text Space 提出GraphText以解决图机器学习中的语言理解问题 large language model
15 Zero-Shot Continuous Prompt Transfer: Generalizing Task Semantics Across Language Models 提出零-shot连续提示迁移方法以解决语言模型间迁移性问题 large language model
16 LLM Lies: Hallucinations are not Bugs, but Features as Adversarial Examples 提出对抗性示例视角下的幻觉攻击方法以解决LLM信任问题 large language model ✅
17 BTR: Binary Token Representations for Efficient Retrieval Augmented Language Models 提出二进制令牌表示以提高检索增强语言模型的效率 large language model
18 SPELL: Semantic Prompt Evolution based on a LLM 提出SPELL方法以优化文本提示的生成 large language model
19 NarrativePlay: Interactive Narrative Understanding 提出NarrativePlay以解决互动叙事理解问题 large language model
20 Automated Evaluation of Classroom Instructional Support with LLMs and BoWs: Connecting Global Predictions to Specific Feedback 利用大语言模型和词袋模型自动评估课堂教学支持 large language model
21 Do Compressed LLMs Forget Knowledge? An Experimental Study with Practical Implications 提出推理重定向方法以恢复压缩LLM的知识性能 large language model
22 Natural Language Models for Data Visualization Utilizing nvBench Dataset 提出自然语言模型以解决数据可视化命令生成问题 large language model
23 Closing the Curious Case of Neural Text Degeneration 提出改进的截断采样策略以解决文本生成中的退化问题 large language model
24 Label Supervised LLaMA Finetuning 提出标签监督的LLaMA微调方法以提升分类任务性能 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (5 篇)

#题目一句话要点标签🔗⭐
25 Towards LogiGLUE: A Brief Survey and A Benchmark for Analyzing Logical Reasoning Capabilities of Language Models 提出LogiGLUE基准以分析语言模型的逻辑推理能力 distillation large language model chain-of-thought
26 Probing the Multi-turn Planning Capabilities of LLMs via 20 Question Games 提出实体推理游戏以评估LLMs的多轮规划能力 reinforcement learning behavior cloning large language model
27 Tool-Augmented Reward Modeling 提出工具增强的奖励建模以解决传统方法的局限性 reinforcement learning RLHF large language model ✅
28 Enabling Language Models to Implicitly Learn Self-Improvement 提出隐式自我改进框架以提升语言模型响应质量 reinforcement learning RLHF large language model
29 UltraFeedback: Boosting Language Models with Scaled AI Feedback 提出UltraFeedback以解决人类反馈获取瓶颈问题 reinforcement learning large language model ✅

⬅️ 返回 cs.CL 首页 · 🏠 返回主页