cs.CL（2024-08-15）

📊 共 20 篇论文 | 🔗 5 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (18 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (2 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (18 篇)

#	题目	一句话要点	标签	🔗	⭐
1	Dynamic Adaptive Optimization for Effective Sentiment Analysis Fine-Tuning on Large Language Models	提出动态自适应优化模块，提升大型语言模型在情感分析微调中的性能	large language model
2	FactorLLM: Factorizing Knowledge via Mixture of Experts for Large Language Models	FactorLLM：通过混合专家模型分解知识，提升大语言模型效率。	large language model	✅
3	P/D-Serve: Serving Disaggregated Large Language Model at Scale	P/D-Serve：大规模解耦LLM服务系统，优化预填充和解码性能	large language model
4	Enhancing Large Language Model-based Speech Recognition by Contextualization for Rare and Ambiguous Words	提出基于LLM的语音识别系统，通过上下文关键词提示提升稀有和歧义词识别。	large language model
5	ArabLegalEval: A Multitask Benchmark for Assessing Arabic Legal Knowledge in Large Language Models	ArabLegalEval：用于评估大型语言模型阿拉伯语法律知识的多任务基准	large language model	✅
6	Predicting Lung Cancer Patient Prognosis with Large Language Models	利用大型语言模型预测肺癌患者预后，无需额外患者数据	large language model
7	Inductive Learning of Logical Theories with LLMs: An Expressivity-Graded Analysis	提出一种新方法，通过形式推理引擎反馈分析LLM在逻辑理论归纳中的能力与局限性。	large language model symbolic grounding
8	FuseChat: Knowledge Fusion of Chat Models	FuseChat：通过轻量级持续训练融合多个聊天模型知识，提升性能并降低成本。	large language model instruction following	✅
9	Hermes 3 Technical Report	Hermes 3：一个具备卓越推理和创造能力的通用指令及工具使用模型	large language model
10	Zero-Shot Learning and Key Points Are All You Need for Automated Fact-Checking	提出基于零样本学习和关键点的ZSL-KeP框架，用于自动化事实核查。	large language model
11	Towards Realistic Synthetic User-Generated Content: A Scaffolding Approach to Generating Online Discussions	提出多步骤生成框架以创建真实合成用户生成内容	large language model
12	ScalingFilter: Assessing Data Quality through Inverse Utilization of Scaling Laws	提出ScalingFilter，通过缩放律逆向利用评估数据质量，消除参考数据集偏差。	large language model
13	The ShareLM Collection and Plugin: Contributing Human-Model Chats for the Benefit of the Community	提出ShareLM数据集与插件，促进人机对话数据共享，助力开源社区模型发展。	large language model
14	Covert Bias: The Severity of Social Views' Unalignment in Language Models Towards Implicit and Explicit Opinion	揭示语言模型中隐性偏见：社会观点不一致对隐性和显性意见的影响	large language model
15	KOALA: Enhancing Speculative Decoding for LLM via Multi-Layer Draft Heads with Adversarial Learning	KOALA：通过对抗学习的多层Draft Head增强LLM的推测解码	large language model
16	I-SHEEP: Self-Alignment of LLM from Scratch through an Iterative Self-Enhancement Paradigm	I-SHEEP：提出一种迭代自增强范式，实现LLM从零开始的持续自对齐	large language model
17	Leveraging Web-Crawled Data for High-Quality Fine-Tuning	利用网络爬取数据进行高质量微调，提升特定领域大语言模型性能。	large language model
18	Coupling without Communication and Drafter-Invariant Speculative Decoding	提出基于Gumbel采样的无通信耦合方法，提升推测解码性能	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	🔗	⭐
19	MIDAS: Multi-level Intent, Domain, And Slot Knowledge Distillation for Multi-turn NLU	提出MIDAS，利用多层次知识蒸馏提升多轮对话NLU性能	distillation large language model	✅
20	DeepSeek-Prover-V1.5: Harnessing Proof Assistant Feedback for Reinforcement Learning and Monte-Carlo Tree Search	DeepSeek-Prover-V1.5：利用证明助手反馈进行强化学习和蒙特卡洛树搜索，提升定理证明能力。	reinforcement learning

⬅️ 返回 cs.CL 首页 · 🏠 返回主页