cs.CL（2024-03-28）

📊 共 27 篇论文 | 🔗 7 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (18 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (6 🔗1) 支柱一：机器人控制 (Robot Control) (1 🔗1) 支柱三：空间感知与语义 (Perception & Semantics) (1 🔗1) 支柱六：视频提取与匹配 (Video Extraction) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (18 篇)

#题目一句话要点标签🔗⭐
1 Mitigating Misleading Chain-of-Thought Reasoning with Selective Filtering 提出选择性过滤推理器以解决链式推理失效问题 large language model chain-of-thought ✅
2 Make Large Language Model a Better Ranker 提出Aligned Listwise Ranking Objectives以解决LLM推荐系统排名问题 large language model
3 Dataverse: Open-Source ETL (Extract, Transform, Load) Pipeline for Large Language Models 提出Dataverse以解决大规模语言模型数据处理挑战 large language model
4 MineLand: Simulating Large-Scale Multi-Agent Interactions with Limited Multimodal Senses and Physical Needs 提出MineLand以解决多智能体交互的生态有效性问题 multimodal ✅
5 Streamlining Redundant Layers to Compress Large Language Models 提出LLM-Streamline以压缩大型语言模型中的冗余层 large language model ✅
6 HeGTa: Leveraging Heterogeneous Graph-enhanced Large Language Models for Few-shot Complex Table Understanding 提出HGT框架以解决复杂表格理解中的少样本问题 large language model
7 WaterJudge: Quality-Detection Trade-off when Watermarking Large Language Models 提出WaterJudge以解决水印对大语言模型质量影响的问题 large language model
8 Large Language Models Struggle with Unreasonability in Math Problems 提出不合理数学问题基准以评估大语言模型的推理能力 large language model
9 Code Comparison Tuning for Code Large Language Models 提出代码比较调优方法以解决代码错误检测问题 large language model
10 MATEval: A Multi-Agent Discussion Framework for Advancing Open-Ended Text Evaluation 提出MATEval框架以解决开放式文本评估中的不确定性问题 large language model chain-of-thought
11 MFORT-QA: Multi-hop Few-shot Open Rich Table Question Answering 提出MFORT-QA以解决复杂表格问答问题 large language model chain-of-thought
12 JDocQA: Japanese Document Question Answering Dataset for Generative Language Models 提出JDocQA数据集以解决日文文档问答问题 large language model multimodal
13 MUGC: Machine Generated versus User Generated Content Detection 提出MUGC以解决机器生成与用户生成内容的检测问题 large language model
14 Top Leaderboard Ranking = Top Coding Proficiency, Always? EvoEval: Evolving Coding Benchmarks via LLM 提出EvoEval以解决现有编码基准的局限性 instruction following ✅
15 FACTOID: FACtual enTailment fOr hallucInation Detection 提出FACTOID以解决LLM生成内容的幻觉检测问题 large language model
16 Checkpoint Merging via Bayesian Optimization in LLM Pretraining 通过贝叶斯优化实现LLM预训练中的检查点合并 large language model
17 Ungrammatical-syntax-based In-context Example Selection for Grammatical Error Correction 提出基于不合语法的上下文示例选择策略以解决语法错误纠正问题 large language model
18 Learning From Correctness Without Prompting Makes LLM Efficient Reasoner 提出自我纠正推理框架以提升大型语言模型推理效率 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (6 篇)

#题目一句话要点标签🔗⭐
19 Mixed Preference Optimization: Reinforcement Learning with Data Selection and Better Reference Model 提出混合偏好优化方法以解决大语言模型对齐问题 reinforcement learning RLHF DPO
20 Disentangling Length from Quality in Direct Preference Optimization 提出一种正则化策略以解决DPO中的长度偏见问题 reinforcement learning RLHF DPO
21 Jamba: A Hybrid Transformer-Mamba Language Model 提出Jamba模型以提升长文本处理能力 Mamba large language model
22 Fine-Tuning Language Models with Reward Learning on Policy 提出基于策略的奖励学习以优化语言模型 reinforcement learning RLHF large language model ✅
23 sDPO: Don't Use Your Data All at Once 提出sDPO方法以优化大语言模型的对齐训练 DPO direct preference optimization large language model
24 GOLD: Generalized Knowledge Distillation via Out-of-Distribution-Guided Language Data Generation 提出GOLD框架以解决语言模型蒸馏中的数据分布偏差问题 distillation

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
25 TableLLM: Enabling Tabular Data Manipulation by LLMs in Real Office Usage Scenarios 提出TableLLM以解决办公场景中的表格数据处理问题 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🔗⭐
26 Semantic Map-based Generation of Navigation Instructions 提出基于语义地图的导航指令生成方法以解决现有方法的局限性 semantic map ✅

🔬 支柱六：视频提取与匹配 (Video Extraction) (1 篇)

#题目一句话要点标签🔗⭐
27 J-CRe3: A Japanese Conversation Dataset for Real-world Reference Resolution 提出J-CRe3数据集以解决日语对话中的现实参考解析问题 egocentric multimodal

⬅️ 返回 cs.CL 首页 · 🏠 返回主页