cs.CL（2024-12-11）

📊 共 24 篇论文 | 🔗 6 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (19 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (4 🔗1) 支柱一：机器人控制 (Robot Control) (1 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (19 篇)

#题目一句话要点标签🔗⭐
1 EmoVerse: Exploring Multimodal Large Language Models for Sentiment and Emotion Understanding 提出EmoVerse：用于情感理解的多模态大语言模型 large language model multimodal ✅
2 Multimodal Latent Language Modeling with Next-Token Diffusion 提出LatentLM，通过下一token扩散的潜在语言建模统一处理多模态生成与理解任务。 large language model multimodal
3 Large Language Models Still Face Challenges in Multi-Hop Reasoning with External Knowledge 大型语言模型在外部知识多跳推理中仍面临挑战 large language model chain-of-thought
4 Performance of a large language model-Artificial Intelligence based chatbot for counseling patients with sexually transmitted infections and genital diseases Otiz：基于大型语言模型的性传播感染咨询AI聊天机器人 large language model
5 Advancing Single and Multi-task Text Classification through Large Language Model Fine-tuning 通过微调大型语言模型提升单任务和多任务文本分类性能 large language model
6 Assessing Personalized AI Mentoring with Large Language Models in the Computing Field 评估大型语言模型在计算机领域个性化AI指导中的应用 large language model
7 Exploiting the Index Gradients for Optimization-Based Jailbreaking on Large Language Models 提出MAGIC方法，利用索引梯度加速大语言模型优化对抗攻击，提升越狱效率。 large language model ✅
8 TECO: Improving Multimodal Intent Recognition with Text Enhancement through Commonsense Knowledge Extraction TECO：通过常识知识抽取增强文本，提升多模态意图识别性能 multimodal
9 Concept Bottleneck Large Language Models 提出概念瓶颈大语言模型以解决可解释性不足问题 large language model ✅
10 NyayaAnumana & INLegalLlama: The Largest Indian Legal Judgment Prediction Dataset and Specialized Language Model for Enhanced Decision Analysis 提出NyayaAnumana印度法律判决预测数据集与INLegalLlama专用语言模型，提升决策分析。 large language model
11 What Makes In-context Learning Effective for Mathematical Reasoning: A Theoretical Analysis 理论分析揭示上下文学习在数学推理中有效性的关键，并提出LMS3选择方法 large language model
12 EMS: Adaptive Evict-then-Merge Strategy for Head-wise KV Cache Compression Based on Global-Local Importance 提出EMS：基于全局-局部重要性的自适应Evict-then-Merge KV缓存压缩策略 large language model
13 SweetieChat: A Strategy-Enhanced Role-playing Framework for Diverse Scenarios Handling Emotional Support Agent 提出SweetieChat框架，通过策略增强的角色扮演提升情感支持Agent的对话能力。 large language model
14 Code LLMs: A Taxonomy-based Survey 构建代码大语言模型分类体系，综述其在自然语言-编程语言桥接中的应用与挑战 large language model
15 LCFO: Long Context and Long Form Output Dataset and Benchmarking 提出LCFO基准，用于评估长文本摘要和摘要扩展能力，并分析了现有LLM的性能。 large language model
16 Rethinking Comprehensive Benchmark for Chart Understanding: A Perspective from Scientific Literature 提出SCI-CQA基准，用于更全面评估多模态模型在科学文献图表理解方面的能力。 multimodal
17 Multilingual LLMs Inherently Reward In-Language Time-Sensitive Semantic Alignment for Low-Resource Languages 提出CLiTSSA，提升多语言LLM在低资源语言时间敏感语义对齐任务上的性能。 large language model
18 Imitate Before Detect: Aligning Machine Stylistic Preference for Machine-Revised Text Detection 提出Imitate Before Detect方法，提升机器润色文本的检测性能 large language model
19 TinyThinker: Distilling Reasoning through Coarse-to-Fine Knowledge Internalization with Self-Reflection TinyThinker：通过粗细粒度知识内化和自反思蒸馏小型模型推理能力 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🔗⭐
20 jina-clip-v2: Multilingual Multimodal Embeddings for Text and Images 提出jina-clip-v2，通过多任务多阶段对比学习提升多语言多模态文本和图像的嵌入效果。 contrastive learning multimodal ✅
21 SmolTulu: Higher Learning Rate to Batch Size Ratios Can Lead to Better Reasoning in SLMs SmolTulu：通过调整学习率与批量大小比例提升小型语言模型的推理能力 DPO large language model instruction following
22 Bridging Relevance and Reasoning: Rationale Distillation in Retrieval-Augmented Generation 提出RADIO框架，通过理由蒸馏对检索增强生成中的重排序器进行偏好对齐。 distillation large language model
23 Learning to Reason via Self-Iterative Process Feedback for Small Language Models 提出基于自迭代过程反馈的小语言模型推理学习方法 distillation large language model

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
24 Detecting Conversational Mental Manipulation with Intent-Aware Prompting 提出Intent-Aware Prompting方法，利用LLM检测对话中的精神操控行为 manipulation large language model ✅

⬅️ 返回 cs.CL 首页 · 🏠 返回主页