cs.CL（2025-01-22）

📊 共 17 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (12 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (5 🔗1)

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

#题目一句话要点标签🔗⭐
1 Refining Input Guardrails: Enhancing LLM-as-a-Judge Efficiency Through Chain-of-Thought Fine-Tuning and Alignment 通过思维链微调与对齐，提升LLM作为输入防护栏的效率 large language model chain-of-thought
2 Does Table Source Matter? Benchmarking and Improving Multimodal Scientific Table Understanding and Reasoning 提出MMSci框架，提升多模态大模型在科学表格理解和数值推理上的能力。 large language model multimodal ✅
3 OnionEval: An Unified Evaluation of Fact-conflicting Hallucination for Small-Large Language Models OnionEval：统一评估大小型语言模型中事实冲突性幻觉的框架 large language model chain-of-thought
4 Architectural Fusion Through Contextual Partitioning in Large Language Models: A Novel Approach to Parameterized Knowledge Integration 提出上下文分区方法，通过动态参数分割提升大语言模型的知识集成能力。 large language model
5 WisdomBot: Tuning Large Language Models with Artificial Intelligence Knowledge WisdomBot：利用人工智能知识微调大型语言模型，提升教育领域应用效果 large language model
6 PairJudge RM: Perform Best-of-N Sampling with Knockout Tournament 提出PairJudge RM，结合淘汰赛机制，提升大语言模型Best-of-N采样效果 large language model chain-of-thought
7 Test-Time Preference Optimization: On-the-Fly Alignment via Iterative Textual Feedback 提出测试时偏好优化（TPO），通过迭代文本反馈实现LLM的即时对齐。 large language model instruction following ✅
8 LLMs as Repositories of Factual Knowledge: Limitations and Solutions 提出实体感知微调(ENAF)，提升LLM在时效性事实知识问答中的准确性和一致性 large language model TAMP
9 Open or Closed LLM for Lesser-Resourced Languages? Lessons from Greek 评估开放与封闭LLM以解决希腊语NLP挑战 large language model
10 NExtLong: Toward Effective Long-Context Training without Long Documents NExtLong：通过负样本扩展实现高效的长文本上下文训练，无需依赖长文档。 large language model
11 Efficient Prompt Compression with Evaluator Heads for Long-Context Transformer Inference 提出基于评估头的高效提示压缩方法，加速长文本Transformer推理。 large language model
12 ACEBench: Who Wins the Match Point in Tool Usage? ACEBench：评估LLM工具使用能力的多维度综合基准 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🔗⭐
13 Quantification of Large Language Model Distillation 提出量化大语言模型蒸馏框架，评估模型同质化程度与身份认知偏差。 distillation large language model ✅
14 DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning DeepSeek-R1：通过强化学习激励LLM的推理能力，无需人工标注 reinforcement learning large language model chain-of-thought
15 OpenGenAlign: A Preference Dataset and Benchmark for Trustworthy Reward Modeling in Open-Ended, Long-Context Generation 提出OpenGenAlign，用于开放域长文本生成中可信奖励建模的偏好数据集与基准。 reinforcement learning large language model instruction following
16 Training Dialogue Systems by AI Feedback for Improving Overall Dialogue Impression 利用AI反馈训练对话系统，提升整体对话体验 reinforcement learning large language model
17 Extracting General-use Transformers for Low-resource Languages via Knowledge Distillation 提出基于知识蒸馏的通用Transformer提取方法，用于低资源语言场景。 distillation

⬅️ 返回 cs.CL 首页 · 🏠 返回主页