cs.CL（2025-07-14）

📊 共 22 篇论文 | 🔗 1 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (19) 支柱二：RL算法与架构 (RL & Architecture) (3 🔗1)

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

#题目一句话要点标签🔗⭐
1 Can Multimodal Foundation Models Understand Schematic Diagrams? An Empirical Study on Information-Seeking QA over Scientific Papers 提出MISS-QA基准，评估多模态模型理解科学论文示意图的能力 foundation model multimodal
2 DeepWriter: A Fact-Grounded Multimodal Writing Assistant Based On Offline Knowledge Base DeepWriter：基于离线知识库的事实性多模态写作助手，提升专业领域文档生成质量。 large language model multimodal
3 MultiVox: A Benchmark for Evaluating Voice Assistants for Multimodal Interactions MultiVox：用于评估多模态交互语音助手的新基准 large language model multimodal
4 HKGAI-V1: Towards Regional Sovereign Large Language Model for Hong Kong 提出HKGAI-V1，一个为香港定制的区域主权大语言模型，关注文化和法律对齐。 large language model
5 Enhancing Chain-of-Thought Reasoning with Critical Representation Fine-tuning 提出CRFT：通过关键表征微调增强Chain-of-Thought推理能力 chain-of-thought
6 Cultural Bias in Large Language Models: Evaluating AI Agents through Moral Questionnaires 揭示大语言模型文化偏见：道德问卷评估AI代理的文化价值观 large language model
7 LLMs on Trial: Evaluating Judicial Fairness for Large Language Models 构建JudiFair数据集与评估框架，揭示LLM在司法公平性上的不足 large language model
8 MLAR: Multi-layer Large Language Model-based Robotic Process Automation Applicant Tracking 提出基于多层大语言模型的MLAR，提升机器人流程自动化在招聘申请追踪中的效率。 large language model
9 Fusing Large Language Models with Temporal Transformers for Time Series Forecasting 融合大语言模型与时序Transformer用于时间序列预测 large language model
10 Function Induction and Task Generalization: An Interpretability Study with Off-by-One Addition 通过可解释性分析揭示大语言模型在Off-by-One加法任务中的泛化机制 large language model
11 CodeJudgeBench: Benchmarking LLM-as-a-Judge for Coding Tasks 提出CodeJudgeBench，评估LLM作为代码评判者在代码生成、修复和测试生成任务中的性能。 large language model
12 Can You Detect the Difference? 系统比较扩散模型与自回归模型生成文本的差异，揭示现有检测器的局限性 large language model
13 Retention analysis of edited knowledge after fine-tuning 研究微调对编辑后知识的遗忘效应，并提出增强知识保留的方法 large language model
14 From Words to Proverbs: Evaluating LLMs Linguistic and Cultural Competence in Saudi Dialects with Absher 提出Absher基准，评估LLM在沙特方言中的语言和文化能力 large language model
15 Enhancing Retrieval Augmented Generation with Hierarchical Text Segmentation Chunking 提出基于层级文本分割的RAG增强方法，提升检索信息的语义连贯性与准确性 large language model
16 Using AI to replicate human experimental results: a motion study 利用AI复现人类实验结果：一项基于动作研究的语言学探索 large language model
17 Grammar-Guided Evolutionary Search for Discrete Prompt Optimisation 提出一种基于语法引导进化搜索的离散提示优化方法，提升小模型在复杂任务上的性能。 large language model
18 GeLaCo: An Evolutionary Approach to Layer Compression 提出GeLaCo以解决大语言模型压缩问题 large language model
19 Protective Factor-Aware Dynamic Influence Learning for Suicide Risk Prediction on Social Media 提出保护因素感知的动态影响学习方法，用于社交媒体上的自杀风险预测。 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (3 篇)

#题目一句话要点标签🔗⭐
20 Flexible Feature Distillation for Large Language Models Flex-KD：一种面向大语言模型的灵活特征蒸馏方法，无需参数对齐，提升下游任务性能。 teacher-student distillation large language model
21 Bridging Robustness and Generalization Against Word Substitution Attacks in NLP via the Growth Bound Matrix Approach 提出基于增长界矩阵的正则化方法，提升NLP模型在词替换攻击下的鲁棒性和泛化性 SSM state space model ✅
22 Tiny Reward Models 提出TinyRM，一种小型双向语言模型，用于高效奖励建模。 reinforcement learning RLHF

⬅️ 返回 cs.CL 首页 · 🏠 返回主页