cs.CL（2025-10-18）

📊 共 17 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) (15 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (2)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (15 篇)

#题目一句话要点标签🔗⭐
1 Probing the Hidden Talent of ASR Foundation Models for L2 English Oral Assessment 探索ASR基础模型Whisper在L2英语口语评估中的潜力，超越现有方法。 foundation model multimodal
2 Fine-tuning of Large Language Models for Constituency Parsing Using a Sequence to Sequence Approach 通过序列到序列方法微调大型语言模型，用于成分句法分析 large language model
3 Hallucination Benchmark for Speech Foundation Models 提出SHALLLOW基准，用于评估语音基础模型中的幻觉现象。 foundation model
4 Language over Content: Tracing Cultural Understanding in Multilingual Large Language Models 通过激活路径分析，揭示多语言大模型中的文化理解机制 large language model
5 FrugalPrompt: Reducing Contextual Overhead in Large Language Models via Token Attribution FrugalPrompt：通过Token归因减少大语言模型中的上下文开销 large language model ✅
6 Utilising Large Language Models for Generating Effective Counter Arguments to Anti-Vaccine Tweets 利用大型语言模型生成针对反疫苗推文的有效反驳论点 large language model
7 Instant Personalized Large Language Model Adaptation via Hypernetwork 提出Profile-to-PEFT框架，通过超网络实现即时个性化大语言模型适配。 large language model
8 When Models Can't Follow: Testing Instruction Adherence Across 256 LLMs 提出一种高效的LLM指令遵循能力评估框架，诊断模型在格式、内容、逻辑等方面的不足。 large language model instruction following
9 Unleashing Diverse Thinking Modes in LLMs through Multi-Agent Collaboration 提出DiMo框架，通过多Agent协作激发LLM多样化思维模式，提升推理能力和可解释性 large language model
10 AI-Generated Text Detection in Low-Resource Languages: A Case Study on Urdu 针对乌尔都语，提出AI生成文本检测框架，有效识别机器生成内容。 large language model
11 ReviewGuard: Enhancing Deficient Peer Review Detection via LLM-Driven Data Augmentation ReviewGuard：利用LLM驱动的数据增强提升缺陷同行评审检测 large language model ✅
12 Check Yourself Before You Wreck Yourself: Selectively Quitting Improves LLM Agent Safety 提出基于“退出”机制的LLM Agent安全策略，显著提升安全性。 large language model
13 TrajSelector: Harnessing Latent Representations for Efficient and Effective Best-of-N in Large Reasoning Model TrajSelector：利用LLM隐空间表征，高效选择最优推理轨迹，提升大模型推理性能。 large language model
14 ATA: A Neuro-Symbolic Approach to Implement Autonomous and Trustworthy Agents 提出ATA神经符号方法，提升自主Agent的可靠性与透明度 large language model
15 Navigating through the hidden embedding space: steering LLMs to improve mental health assessment 提出基于线性变换的LLM引导方法，提升心理健康评估能力 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🔗⭐
16 RAVEN: Robust Advertisement Video Violation Temporal Grounding via Reinforcement Reasoning RAVEN：通过强化推理实现鲁棒的广告视频违规行为时序定位 reinforcement learning large language model multimodal
17 Thinking About Thinking: Evaluating Reasoning in Post-Trained Language Models 评估后训练语言模型推理能力的“思考”能力 DPO large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页