cs.CL（2025-05-03）

📊 共 9 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) (8 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (8 篇)

#	题目	一句话要点	标签	🔗	⭐
1	Adaptive Token Boundaries: Integrating Human Chunking Mechanisms into Multimodal LLMs	提出自适应Token边界的多模态LLM，模拟人类Chunking机制以提升跨模态信息整合能力	large language model multimodal
2	Accelerating Large Language Model Reasoning via Speculative Search	提出SpecSearch，通过推测搜索加速大语言模型推理过程。	large language model
3	A Multimodal Framework for Explainable Evaluation of Soft Skills in Educational Environments	提出一种多模态模糊逻辑框架，用于教育环境中软技能的可解释评估。	multimodal
4	Same evaluation, more tokens: On the effect of input length for machine translation evaluation using Large Language Models	研究表明输入长度影响LLM机器翻译评估，提出FSP和微调方法缓解该问题。	large language model
5	Efficient Shapley Value-based Non-Uniform Pruning of Large Language Models	提出基于Shapley值的非均匀剪枝方法，提升大语言模型剪枝后的性能。	large language model
6	A Survey on Inference Engines for Large Language Models: Perspectives on Optimization and Efficiency	评估大型语言模型推理引擎以提升效率与优化	large language model	✅
7	$\textit{New News}$: System-2 Fine-tuning for Robust Integration of New Knowledge	提出System-2微调方法，提升LLM对新知识的稳健整合能力，缩小微调与上下文学习的差距。	large language model
8	Cannot See the Forest for the Trees: Invoking Heuristics and Biases to Elicit Irrational Choices of LLMs	ICRT框架：利用人类认知偏差诱导大语言模型产生有害内容	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	🔗	⭐
9	High-Fidelity Pseudo-label Generation by Large Language Models for Training Robust Radiology Report Classifiers	提出DeBERTa-RAD框架，利用LLM伪标签和知识蒸馏训练鲁棒的放射报告分类器。	distillation large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页