cs.CL（2024-07-17）

📊 共 28 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) (23 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (5)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (23 篇)

#题目一句话要点标签🔗⭐
1 E5-V: Universal Embeddings with Multimodal Large Language Models E5-V：利用多模态大语言模型实现通用多模态嵌入 large language model multimodal
2 MERLIN: Multimodal Embedding Refinement via LLM-based Iterative Navigation for Text-Video Retrieval-Rerank Pipeline MERLIN：利用LLM迭代导航的多模态嵌入优化文本-视频检索重排序流水线 large language model foundation model multimodal
3 Evaluating Linguistic Capabilities of Multimodal LLMs in the Lens of Few-Shot Learning 评估多模态LLM在小样本学习中的语言能力，关注ICL和CoT提示 large language model multimodal chain-of-thought
4 LMMs-Eval: Reality Check on the Evaluation of Large Multimodal Models 提出LMMS-EVAL框架，解决大模型评测中覆盖率、成本和污染的难题。 foundation model multimodal ✅
5 Beyond Next Token Prediction: Patch-Level Training for Large Language Models 提出Patch-Level训练方法，在不牺牲性能的前提下显著降低大语言模型的训练成本。 large language model ✅
6 A Survey of Prompt Engineering Methods in Large Language Models for Different NLP Tasks 综述Prompt工程在大型语言模型中应用于不同NLP任务的方法 large language model
7 Struct-X: Enhancing Large Language Models Reasoning with Structured Data Struct-X：利用结构化数据增强大语言模型的推理能力 large language model
8 Explainable Biomedical Hypothesis Generation via Retrieval Augmented Generation enabled Large Language Models 提出RUGGED框架，利用RAG-LLM进行可解释的生物医学假设生成，辅助药物发现。 large language model
9 Multimodal Reranking for Knowledge-Intensive Visual Question Answering 提出多模态重排序模块，提升知识密集型视觉问答中知识候选的排序质量。 multimodal
10 Is Sarcasm Detection A Step-by-Step Reasoning Process in Large Language Models? 提出SarcasmCue框架，探索大语言模型中逐步推理对反讽检测的影响 large language model
11 Matryoshka-Adaptor: Unsupervised and Supervised Tuning for Smaller Embedding Dimensions Matryoshka-Adaptor：通过无监督和监督调优，降低LLM Embedding维度并保持性能。 large language model multimodal
12 Steamroller Problems: An Evaluation of LLM Reasoning Capability with Automated Theorem Prover Strategies 评估LLM在自动定理证明策略下的推理能力：基于Steamroller问题的研究 large language model chain-of-thought
13 TurkishMMLU: Measuring Massive Multitask Language Understanding in Turkish 提出TurkishMMLU：首个土耳其语多任务选择题基准，用于评估LLM的理解能力。 large language model chain-of-thought ✅
14 Halu-J: Critique-Based Hallucination Judge 提出Halu-J，一种基于批判的多证据幻觉检测模型，提升LLM生成内容的事实性。 large language model ✅
15 Harnessing the Power of Artificial Intelligence to Vitalize Endangered Indigenous Languages: Technologies and Experiences 利用AI和NLP技术，特别是LLM，促进濒危土著语言的使用和记录。 large language model
16 Sharif-STR at SemEval-2024 Task 1: Transformer as a Regression Model for Fine-Grained Scoring of Textual Semantic Relations 利用RoBERTa微调进行文本语义关系细粒度评分，提升多语言STR性能 large language model
17 AudienceView: AI-Assisted Interpretation of Audience Feedback in Journalism AudienceView：利用大型语言模型辅助记者解读海量受众反馈 large language model
18 Crafting the Path: Robust Query Rewriting for Information Retrieval 提出Crafting the Path结构化查询重写方法，提升信息检索在低资源领域的鲁棒性 large language model
19 Case2Code: Scalable Synthetic Data for Code Generation 提出Case2Code任务，通过大规模合成数据提升代码生成模型性能 large language model
20 Automate or Assist? The Role of Computational Models in Identifying Gendered Discourse in US Capital Trial Transcripts 利用计算模型辅助法律专家识别美国死刑审判中性别歧视性言论 large language model
21 Krutrim LLM: A Novel Tokenization Strategy for Multilingual Indic Languages with Petabyte-Scale Data Processing Krutrim LLM：面向多语种印度语的PB级数据处理与新型分词策略 large language model
22 Navigating the Noisy Crowd: Finding Key Information for Claim Verification 提出EACon框架，通过证据抽象和主张解构提升LLM在声明验证中的性能 large language model
23 The Better Angels of Machine Personality: How Personality Relates to LLM Safety 从人格视角探索LLM安全性：揭示人格特质与安全能力的关联 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🔗⭐
24 MEDFuse: Multimodal EHR Data Fusion with Masked Lab-Test Modeling and Large Language Models 提出MEDFuse以解决多模态电子健康记录数据融合问题 predictive model large language model multimodal
25 PersLLM: A Personified Training Approach for Large Language Models PersLLM：一种用于大型语言模型的人格化训练方法，提升模型在人机交互和多智能体系统中的表现。 DPO large language model chain-of-thought
26 Pre-Trained Foundation Model representations to uncover Breathing patterns in Speech 利用预训练模型表征，从语音中识别呼吸模式以进行呼吸率估计 MAE foundation model
27 Towards Collaborative Intelligence: Propagating Intentions and Reasoning for Multi-Agent Coordination with Large Language Models 提出基于LLM的协作智能框架，通过意图传播提升多智能体协同能力 reinforcement learning large language model
28 Subgraph-Aware Training of Language Models for Knowledge Graph Completion Using Structure-Aware Contrastive Learning 提出SATKGC框架，利用子图感知训练提升语言模型在知识图谱补全任务上的性能 contrastive learning

⬅️ 返回 cs.CL 首页 · 🏠 返回主页