cs.CL（2025-12-16）

📊 共 25 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) (23 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (2)

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

#题目一句话要点标签🔗⭐
1 Parameter Efficient Multimodal Instruction Tuning for Romanian Vision Language Models 提出针对罗马尼亚语视觉语言模型的参数高效多模态指令微调方法 multimodal
2 Multiscale Aggregated Hierarchical Attention (MAHA): A Game Theoretic and Optimization Driven Approach to Efficient Contextual Modeling in Large Language Models 提出多尺度聚合分层注意力（MAHA），高效建模长文本上下文，降低LLM计算复杂度。 large language model
3 Integrating Large Language Models and Knowledge Graphs to Capture Political Viewpoints in News Media 融合大型语言模型与知识图谱以捕捉新闻媒体中的政治观点 large language model
4 JMMMU-Pro: Image-based Japanese Multi-discipline Multimodal Understanding Benchmark via Vibe Benchmark Construction 提出JMMMU-Pro日语多学科多模态理解基准，并提出Vibe基准构建方法。 multimodal
5 Agreement Between Large Language Models and Human Raters in Essay Scoring: A Research Synthesis 综述研究：大型语言模型在作文评分中与人类评分者的一致性分析 large language model
6 VLegal-Bench: Cognitively Grounded Benchmark for Vietnamese Legal Reasoning of Large Language Models 提出VLegal-Bench，用于评估大型语言模型在越南法律推理任务中的能力。 large language model
7 SASQ: Static Activation Scaling for Quantization-Aware Training in Large Language Models SASQ：一种用于大语言模型量化感知训练的静态激活缩放方法 large language model
8 Effect of Document Packing on the Latent Multi-Hop Reasoning Capabilities of Large Language Models 研究文档打包策略对大语言模型多跳推理能力的影响 large language model
9 Inflation Attitudes of Large Language Models 利用大型语言模型模拟通胀预期，揭示其对宏观经济信号的认知能力。 large language model
10 CogMem: A Cognitive Memory Architecture for Sustained Multi-Turn Reasoning in Large Language Models CogMem：一种认知记忆架构，用于大型语言模型中持续的多轮推理 large language model
11 What Affects the Effective Depth of Large Language Models? 研究揭示大语言模型有效深度受限，提出提升层利用率的研究方向 large language model ✅
12 Roles of MLLMs in Visually Rich Document Retrieval for RAG: A Survey 综述MLLM在富视觉文档RAG检索中的应用，分析三种角色及其优劣势。 large language model multimodal
13 Ladder Up, Memory Down: Low-Cost Fine-Tuning With Side Nets Ladder Side Tuning通过轻量级侧网络实现低成本大模型微调，显著降低内存占用。 large language model chain-of-thought
14 Scalable Frameworks for Real-World Audio-Visual Speech Recognition 提出可扩展框架，提升AVSR系统在真实环境下的鲁棒性 foundation model multimodal
15 T5Gemma 2: Seeing, Reading, and Understanding Longer T5Gemma 2：提出一种轻量级多模态长文本理解的Encoder-Decoder模型。 multimodal
16 Incentives or Ontology? A Structural Rebuttal to OpenAI's Hallucination Thesis 挑战OpenAI幻觉理论：Transformer结构性缺陷导致幻觉，而非激励不足 large language model
17 VersatileFFN: Achieving Parameter Efficiency in LLMs via Adaptive Wide-and-Deep Reuse VersatileFFN：通过自适应宽深复用提升LLM的参数效率 large language model ✅
18 C-ing Clearly: Enhanced Binary Code Explanations using C code C-ing Clearly：利用C代码增强LLM对二进制代码的理解，提升代码解释能力 large language model
19 Two CFG Nahuatl for automatic corpora expansion 提出两种CFG Nahuatl方法，用于自动扩展Nawatl语料库 large language model
20 Astraea: A State-Aware Scheduling Engine for LLM-Powered Agents Astraea：面向LLM智能体的状态感知调度引擎，优化端到端延迟 large language model
21 Multilingual and Continuous Backchannel Prediction: A Cross-lingual Study 提出一种多语种连续后通道预测模型，用于研究跨语言的时序行为差异。 zero-shot transfer
22 A Unified Sparse Attention via Multi-Granularity Compression 提出UniSparse以解决长序列自注意力计算瓶颈问题 large language model
23 Structure-Aware Decoding Mechanisms for Complex Entity Extraction with Large-Scale Language Models 提出结构感知解码方法，利用大语言模型解决复杂实体抽取中语义完整性和结构一致性问题。 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (2 篇)

#题目一句话要点标签🔗⭐
24 Internal Reasoning vs. External Control: A Thermodynamic Analysis of Sycophancy in Large Language Models 提出RCA方法以解决大型语言模型中的谄媚问题 RLHF large language model
25 Cross-Tokenizer Likelihood Scoring Algorithms for Language Model Distillation 提出基于BPE递归结构的跨Tokenizer似然评分算法，用于语言模型蒸馏。 distillation

⬅️ 返回 cs.CL 首页 · 🏠 返回主页