cs.CL（2025-03-29）

📊 共 13 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) (12 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (12 篇)

#	题目	一句话要点	标签	🔗	⭐
1	RECALL-MM: A Multimodal Dataset of Consumer Product Recalls for Risk Analysis using Computational Methods and Large Language Models	构建RECALL-MM多模态数据集，利用计算方法和LLM进行消费品召回风险分析。	large language model multimodal
2	Efficient Inference for Large Reasoning Models: A Survey	综述：面向大型推理模型的高效推理方法，旨在缓解token低效问题。	large language model chain-of-thought	✅
3	FReM: A Flexible Reasoning Mechanism for Balancing Quick and Slow Thinking in Long-Context Question Answering	提出FReM，平衡长文本问答中快速与慢速推理，提升复杂问题解答能力	large language model chain-of-thought
4	Evaluating how LLM annotations represent diverse views on contentious topics	评估LLM标注在争议性话题上对不同观点的代表性，揭示潜在偏见来源。	large language model
5	Beyond speculation: Measuring the growing presence of LLM-generated texts in multilingual disinformation	实证研究揭示LLM生成文本在多语种虚假信息中日益增长的存在	large language model
6	The Reasoning-Memorization Interplay in Language Models Is Mediated by a Single Direction	揭示LLM推理与记忆间的线性特征，实现可控的性能提升	large language model
7	LangVAE and LangSpace: Building and Probing for Language Model VAEs	LangVAE：构建并探究基于预训练语言模型的变分自编码器	large language model
8	Cooperative Speech, Semantic Competence, and AI	探讨大型语言模型是否具备合作对话能力及语义能力	large language model
9	The realization of tones in spontaneous spoken Taiwan Mandarin: a corpus-based survey and theory-driven computational modeling	利用语料库和计算模型研究台湾闽南语口语中音调的实现，揭示语义对音调的影响	large language model
10	EventWeave: A Dynamic Framework for Capturing Core and Supporting Events in Dialogue Systems	EventWeave：一种动态框架，用于在对话系统中捕获核心和支持事件，提升上下文理解。	large language model
11	S2MoE: Robust Sparse Mixture of Experts via Stochastic Learning	S2MoE：基于随机学习的鲁棒稀疏专家混合模型，提升模型性能并降低推理成本	large language model
12	Unified Sparse Mixture of Experts	提出统一稀疏专家混合模型(USMoE)，解决传统SMoE路由的局限性。	large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (1 篇)

#	题目	一句话要点	标签	🔗	⭐
13	Beyond Contrastive Learning: Synthetic Data Enables List-wise Training with Multiple Levels of Relevance	利用合成数据进行列表式训练，实现多层次相关性建模，超越对比学习	contrastive learning large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页