cs.CL（2025-07-23）

📊 共 15 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) (11 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (3) 支柱三：空间感知与语义 (Perception & Semantics) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (11 篇)

#题目一句话要点标签🔗⭐
1 Evaluating the Performance of AI Text Detectors, Few-Shot and Chain-of-Thought Prompting Using DeepSeek Generated Text 评估AI文本检测器对DeepSeek生成文本的检测性能，考察少样本和思维链提示的影响 large language model chain-of-thought
2 A Hybrid Early-Exit Algorithm for Large Language Models Based on Space Alignment Decoding (SPADE) 提出SPADE：一种基于空间对齐解码的混合早期退出算法，用于加速大语言模型推理。 large language model
3 The Pluralistic Moral Gap: Understanding Judgment and Value Differences between Humans and Large Language Models 提出道德困境数据集与动态道德剖析方法，提升LLM道德判断与人类价值对齐程度。 large language model
4 Dynamic and Generalizable Process Reward Modeling 提出动态可泛化的过程奖励建模(DG-PRM)，提升LLM在复杂任务中的性能。 large language model
5 Towards Greater Leverage: Scaling Laws for Efficient Mixture-of-Experts Language Models 提出效率杠杆(EL)指标，揭示MoE模型高效扩展的缩放规律 large language model
6 PRGB Benchmark: A Robust Placeholder-Assisted Algorithm for Benchmarking Retrieval-Augmented Generation 提出PRGB基准，用于评估检索增强生成中LLM的文档利用能力。 large language model ✅
7 Who Attacks, and Why? Using LLMs to Identify Negative Campaigning in 18M Tweets across 19 Countries 利用大型语言模型识别19国1800万条推文中的负面竞选活动 large language model
8 MultiNRC: A Challenging and Native Multilingual Reasoning Evaluation Benchmark for LLMs MultiNRC：一个用于评估LLM多语言推理能力的具挑战性的原生基准 large language model
9 Hierarchical Memory for High-Efficiency Long-Term Reasoning in LLM Agents 提出层级记忆（H-MEM）架构，提升LLM Agent长期推理效率 large language model
10 Tab-MIA: A Benchmark Dataset for Membership Inference Attacks on Tabular Data in LLMs Tab-MIA：用于评估LLM在表格数据上成员推断攻击的基准数据集 large language model
11 SKA-Bench: A Fine-Grained Benchmark for Evaluating Structured Knowledge Understanding of LLMs SKA-Bench：用于评估LLM结构化知识理解能力的细粒度基准测试 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🔗⭐
12 Shop-R1: Rewarding LLMs to Simulate Human Behavior in Online Shopping via Reinforcement Learning 提出Shop-R1框架以增强LLM在在线购物中的人类行为模拟能力 reinforcement learning large language model
13 CogDual: Enhancing Dual Cognition of LLMs via Reinforcement Learning with Implicit Rule-Based Rewards 提出CogDual，通过强化学习和隐式规则奖励增强LLM的角色扮演双重认知能力 reinforcement learning large language model
14 Megrez2 Technical Report Megrez2：一种轻量级高性能语言模型架构，优化设备原生部署。 reinforcement learning instruction following

🔬 支柱三：空间感知与语义 (Perception & Semantics) (1 篇)

#题目一句话要点标签🔗⭐
15 VeriMinder: Mitigating Analytical Vulnerabilities in NL2SQL VeriMinder：缓解NL2SQL中分析漏洞的交互式系统 semantic mapping semantic map

⬅️ 返回 cs.CL 首页 · 🏠 返回主页