cs.CL（2025-05-07）

📊 共 23 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) 支柱二：RL算法与架构 (RL & Architecture) (5) 支柱四：生成式动作 (Generative Motion) (1) 支柱一：机器人控制 (Robot Control) (1)

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

#题目一句话要点标签🔗⭐
1 Enhancing Granular Sentiment Classification with Chain-of-Thought Prompting in Large Language Models 利用思维链提示增强大语言模型在细粒度情感分类中的性能 large language model chain-of-thought
2 Fine-Tuning Large Language Models and Evaluating Retrieval Methods for Improved Question Answering on Building Codes 针对建筑规范问答，提出微调大语言模型并评估检索方法以提升性能。 large language model
3 Personalized Risks and Regulatory Strategies of Large Language Models in Digital Advertising 提出基于BERT的个性化广告推荐模型，兼顾用户隐私保护与广告效果提升 large language model
4 Large Means Left: Political Bias in Large Language Models Increases with Their Number of Parameters 大型语言模型参数越多，政治偏见越严重：偏向左翼立场 large language model
5 Large Language Models are often politically extreme, usually ideologically inconsistent, and persuasive even in informational contexts 揭示大语言模型政治立场极端化、意识形态不一致及信息传播中的说服性 large language model
6 LLM-Independent Adaptive RAG: Let the Question Speak for Itself 提出LLM无关的自适应RAG，通过问题本身决定是否检索 large language model
7 Reward-SQL: Boosting Text-to-SQL via Stepwise Reasoning and Process-Supervised Rewards Reward-SQL：通过逐步推理和过程监督奖励提升Text-to-SQL性能 large language model
8 A Tale of Two Identities: An Ethical Audit of Human and AI-Crafted Personas 通过伦理审计揭示LLM生成人物角色中的种族身份偏见与刻板印象 large language model
9 YABLoCo: Yet Another Benchmark for Long Context Code Generation YABLoCo：面向长上下文代码生成的全新基准测试集 large language model
10 REVEAL: Multi-turn Evaluation of Image-Input Harms for Vision LLM 提出REVEAL框架，用于多轮对话中图像输入型视觉语言模型的有害性评估。 large language model
11 Advancing and Benchmarking Personalized Tool Invocation for LLMs 提出PTool框架与PTBench基准，用于评估和提升LLM的个性化工具调用能力 large language model ✅
12 Osiris: A Lightweight Open-Source Hallucination Detection System Osiris：轻量级开源幻觉检测系统，提升RAG系统可靠性 large language model
13 Red Teaming the Mind of the Machine: A Systematic Evaluation of Prompt Injection and Jailbreak Vulnerabilities in LLMs 系统性评估LLM的Prompt注入和越狱漏洞，提出分层缓解策略 large language model
14 Pangu Ultra MoE: How to Train Your Big MoE on Ascend NPUs 提出盘古 Ultra MoE 模型，探索在昇腾 NPU 上训练千亿级稀疏 MoE 大模型的有效方法。 large language model
15 Miipher-2: A Universal Speech Restoration Model for Million-Hour Scale Data Restoration Miipher-2：面向百万小时数据修复的通用语音恢复模型 large language model
16 Benchmarking LLMs' Swarm intelligence SwarmBench：评估LLM在严格群体智能约束下涌现协同能力的基准 large language model ✅

🔬 支柱二：RL算法与架构 (RL & Architecture) (5 篇)

#题目一句话要点标签🔗⭐
17 OBLIVIATE: Robust and Practical Machine Unlearning for Large Language Models 提出OBLIVIATE框架以解决大语言模型中的数据遗忘问题 distillation large language model
18 The Aloe Family Recipe for Open and Specialized Healthcare LLMs Aloe家族开源医疗LLM：优化数据处理与训练，提升安全性和有效性 DPO direct preference optimization large language model
19 HiPerRAG: High-Performance Retrieval Augmented Generation for Scientific Insights HiPerRAG：面向科学洞见的高性能检索增强生成框架 contrastive learning large language model multimodal
20 SOAEsV2-7B/72B: Full-Pipeline Optimization for State-Owned Enterprise LLMs via Continual Pre-Training, Domain-Progressive SFT and Distillation-Enhanced Speculative Decoding SOAEsV2-7B/72B：面向国资企业的全流程优化大语言模型 distillation large language model
21 ZeroSearch: Incentivize the Search Capability of LLMs without Searching ZeroSearch：一种无需真实搜索即可提升LLM搜索能力的新型强化学习框架 reinforcement learning large language model

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
22 AI-Generated Fall Data: Assessing LLMs and Diffusion Model for Wearable Fall Detection 利用AI生成跌倒数据：评估LLM和扩散模型在可穿戴设备跌倒检测中的应用 text-to-motion large language model

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
23 Joint Detection of Fraud and Concept Drift inOnline Conversations with LLM-Assisted Judgment 提出LLM辅助的欺诈和概念漂移联合检测框架，用于在线对话场景 manipulation large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页