cs.CL（2023-10-04）

📊 共 22 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) (17 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (4) 支柱三：空间感知与语义 (Perception & Semantics) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (17 篇)

#题目一句话要点标签🔗⭐
1 On the Performance of Multimodal Language Models 提出多模态指令调优方法以提升语言模型的任务泛化能力 large language model multimodal
2 Large Language Model Cascades with Mixture of Thoughts Representations for Cost-efficient Reasoning 提出LLM级联方法以降低推理任务成本 large language model chain-of-thought
3 How FaR Are Large Language Models From Agents with Theory-of-Mind? 提出Thinking for Doing评估框架以解决LLMs的行动推理问题 large language model chain-of-thought
4 Retrieval meets Long Context Large Language Models 提出检索增强方法以优化长上下文大语言模型的性能 large language model
5 Assessing Large Language Models on Climate Information 提出气候信息评估框架以提升大语言模型表现 large language model
6 Integrating UMLS Knowledge into Large Language Models for Medical Question Answering 提出基于UMLS的增强型LLM框架以改善医疗问答质量 large language model
7 MetaTool Benchmark for Large Language Models: Deciding Whether to Use Tools and Which to Use 提出MetaTool基准以评估大型语言模型的工具使用能力 large language model ✅
8 Multimodal Question Answering for Unified Information Extraction 提出多模态问答框架以统一信息提取任务 multimodal
9 From Words to Watts: Benchmarking the Energy Costs of Large Language Model Inference 基于能耗基准测试研究大语言模型推理的能量成本 large language model
10 A Survey of GPT-3 Family Large Language Models Including ChatGPT and GPT-4 综述GPT-3家族大语言模型的研究进展与未来方向 large language model
11 DOMINO: A Dual-System for Multi-step Visual Language Reasoning 提出双系统以解决多步骤视觉语言推理问题 large language model multimodal
12 DQ-LoRe: Dual Queries with Low Rank Approximation Re-ranking for In-Context Learning 提出DQ-LoRe框架以自动选择示例以提升LLM推理能力 large language model chain-of-thought ✅
13 NOLA: Compressing LoRA using Linear Combination of Random Basis 提出NOLA以解决LoRA参数下限问题 large language model ✅
14 Shadow Alignment: The Ease of Subverting Safely-Aligned Language Models 提出影子对齐攻击以揭示安全对齐语言模型的脆弱性 large language model
15 Can Language Models Employ the Socratic Method? Experiments with Code Debugging 提出自动化苏格拉底教学法以辅助编程调试 chain-of-thought ✅
16 JsonTuning: Towards Generalizable, Robust, and Controllable Instruction Tuning 提出JsonTuning以解决现有指令调优的通用性和可控性问题 large language model
17 AGIR: Automating Cyber Threat Intelligence Reporting with Natural Language Generation 提出AGIR以解决网络威胁情报报告自动化问题 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (4 篇)

#题目一句话要点标签🔗⭐
18 $\mathcal{B}$-Coder: Value-Based Deep Reinforcement Learning for Program Synthesis 提出$ ext{B}$-Coder以解决程序合成中的价值优化问题 reinforcement learning deep reinforcement learning reward design
19 Multimodal Prompt Transformer with Hybrid Contrastive Learning for Emotion Recognition in Conversation 提出多模态提示变换器以解决情感识别中的噪声与标签稀缺问题 contrastive learning multimodal
20 CITING: Large Language Models Create Curriculum for Instruction Tuning 提出CITING方法以解决大语言模型指令调优中的人工瓶颈问题 RLHF large language model
21 I$^2$KD-SLU: An Intra-Inter Knowledge Distillation Framework for Zero-Shot Cross-Lingual Spoken Language Understanding 提出I$^2$KD-SLU框架以解决零样本跨语言口语理解问题 distillation

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

#题目一句话要点标签🔗⭐
22 Discovering Knowledge-Critical Subnetworks in Pretrained Language Models 提出多目标可微掩蔽方案以发现预训练语言模型中的知识关键子网络 implicit representation

⬅️ 返回 cs.CL 首页 · 🏠 返回主页