cs.CL（2024-12-30）

📊 共 19 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (16 🔗2) 支柱三：空间感知与语义 (Perception & Semantics) (1) 支柱八：物理动画 (Physics-based Animation) (1) 支柱二：RL算法与架构 (RL & Architecture) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (16 篇)

#题目一句话要点标签🔗⭐
1 Distributed Mixture-of-Agents for Edge Inference with Large Language Models 提出基于分布式混合Agent的边缘LLM推理框架，提升推理质量。 large language model ✅
2 DoTA: Weight-Decomposed Tensor Adaptation for Large Language Models 提出DoTA以解决大语言模型微调中的初始化问题 large language model
3 Facilitating large language model Russian adaptation with Learned Embedding Propagation 提出LEP方法，通过学习嵌入传播实现大语言模型俄语高效适配，无需指令微调。 large language model
4 KARPA: A Training-free Method of Adapting Knowledge Graph as References for Large Language Model's Reasoning Path Aggregation KARPA：一种免训练的知识图谱推理路径聚合方法，提升大语言模型推理能力 large language model
5 Measuring Large Language Models Capacity to Annotate Journalistic Sourcing 评估大语言模型在新闻报道溯源标注能力，构建新闻伦理基准 large language model
6 HumanEval Pro and MBPP Pro: Evaluating Large Language Models on Self-invoking Code Generation 提出自调用代码生成任务及HumanEval Pro等基准，评估LLM的递进推理能力。 large language model
7 Enhancing Multimodal Emotion Recognition through Multi-Granularity Cross-Modal Alignment 提出多粒度跨模态对齐框架MGCMA，提升多模态情感识别性能。 multimodal
8 Knowledge Editing for Large Language Model with Knowledge Neuronal Ensemble 提出知识神经元集成(KNE)方法，提升大语言模型知识编辑的准确性和效率。 large language model
9 Enhancing Annotated Bibliography Generation with LLM Ensembles 提出基于LLM集成的方法，提升带注释的文献目录生成质量。 large language model
10 The Text Classification Pipeline: Starting Shallow going Deeper 综述文本分类流程：从浅层模型到深度学习方法 large language model
11 Verbosity-Aware Rationale Reduction: Effective Reduction of Redundant Rationale via Principled Criteria 提出基于冗余度的句子级推理精简框架，有效降低大语言模型推理成本。 large language model
12 Disentangling Preference Representation and Text Generation for Efficient Individual Preference Alignment 提出个性化偏好对齐方法以提高LLM交互效率 large language model
13 Attributing Culture-Conditioned Generations to Pretraining Corpora 提出MEMOed框架，分析预训练语料库对文化条件生成中文化偏见的影响 large language model
14 Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs 针对o1类LLM的过度思考问题，提出自训练策略优化推理效率。 chain-of-thought
15 Enhancing AI Safety Through the Fusion of Low Rank Adapters 利用LoRA融合提升大语言模型AI安全性，降低有害回复率 large language model
16 Align Attention Heads Before Merging Them: An Effective Way for Converting MHA to GQA 提出一种高效方法将多头注意力转换为分组查询注意力 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🔗⭐
17 GePBench: Evaluating Fundamental Geometric Perception for Multimodal Large Language Models GePBench：评估多模态大语言模型中的几何感知能力 scene understanding spatial relationship large language model

🔬 支柱八：物理动画 (Physics-based Animation) (1 篇)

#题目一句话要点标签🔗⭐
18 Two-component spatiotemporal template for activation-inhibition of speech in ECoG 提出基于时空模板的双分量模型，用于脑皮层电活动中语音激活与抑制的分析 spatiotemporal

🔬 支柱二：RL算法与架构 (RL & Architecture) (1 篇)

#题目一句话要点标签🔗⭐
19 Temporal reasoning for timeline summarisation in social media 提出基于知识蒸馏的时间推理方法，提升社交媒体时间线摘要质量 distillation large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页