cs.CL（2024-05-31）

📊 共 25 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) (16 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (8 🔗3) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (16 篇)

#题目一句话要点标签🔗⭐
1 Preemptive Answer "Attacks" on Chain-of-Thought Reasoning 揭示预设答案对CoT推理的攻击，并提出缓解措施 large language model chain-of-thought
2 LACIE: Listener-Aware Finetuning for Confidence Calibration in Large Language Models LACIE：面向大语言模型置信度校准的、考虑听众的微调方法 large language model
3 OR-Bench: An Over-Refusal Benchmark for Large Language Models 提出OR-Bench，用于评估和提升大型语言模型的过度拒绝问题。 large language model ✅
4 Don't Buy it! Reassessing the Ad Understanding Abilities of Contrastive Multimodal Models 揭示对比多模态模型在广告理解中利用启发式线索的局限性，并提出TRADE评估集。 multimodal ✅
5 Leveraging Large Language Models for Entity Matching 利用大型语言模型解决实体匹配问题 large language model
6 GAMedX: Generative AI-based Medical Entity Data Extractor Using Large Language Models GAMedX：利用大型语言模型进行医疗实体数据提取的生成式AI方法 large language model
7 Open Ko-LLM Leaderboard: Evaluating Large Language Models in Korean with Ko-H5 Benchmark 构建Open Ko-LLM排行榜与Ko-H5基准，促进韩语LLM的评估与发展 large language model
8 Large Language Models: A New Approach for Privacy Policy Analysis at Scale 利用大型语言模型高效分析大规模隐私政策，降低成本并提升准确率 large language model
9 Joint Embeddings for Graph Instruction Tuning 提出基于图嵌入的指令调优方法，增强LLM的图理解能力 large language model multimodal instruction following
10 Passage-specific Prompt Tuning for Passage Reranking in Question Answering with Large Language Models 提出段落特定Prompt Tuning方法，提升大语言模型在开放域问答中的段落重排序性能 large language model
11 DAFNet: Dynamic Auxiliary Fusion for Sequential Model Editing in Large Language Models 提出DAFNet，用于解决大语言模型中持续出现的知识错误编辑问题。 large language model
12 clembench-2024: A Challenging, Dynamic, Complementary, Multilingual Benchmark and Underlying Flexible Framework for LLMs as Multi-Action Agents Clembench-2024：用于评估LLM多动Agent能力的高挑战性、动态、互补、多语言基准测试框架 large language model instruction following
13 Unveiling the Lexical Sensitivity of LLMs: Combinatorial Optimization for Prompt Enhancement 提出COPLE框架，通过组合优化提升LLM对指令词汇变化的鲁棒性。 large language model
14 That's Optional: A Contemporary Exploration of "that" Omission in English Subordinate Clauses 利用信息熵优化语言模型，研究英语从句中“that”省略现象 large language model
15 DORY: Deliberative Prompt Recovery for LLM DORY：利用概率不确定性进行大语言模型提示词恢复，实现SOTA large language model
16 FineRadScore: A Radiology Report Line-by-Line Evaluation Technique Generating Corrections with Severity Scores 提出FineRadScore，一种基于LLM的胸部X光报告逐行评估与纠错方法 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (8 篇)

#题目一句话要点标签🔗⭐
17 LLM-ESR: Large Language Models Enhancement for Long-tailed Sequential Recommendation 提出LLM-ESR框架，利用大语言模型增强长尾序列推荐系统性能 distillation large language model ✅
18 Learning to Clarify: Multi-turn Conversations with Action-Based Contrastive Self-Training 提出基于动作对比自训练的ACT方法，提升LLM在多轮对话中澄清用户意图的能力。 policy learning DPO direct preference optimization
19 Direct Alignment of Language Models via Quality-Aware Self-Refinement 提出质量感知自精炼方法，直接对齐语言模型，提升DPO训练效果 reinforcement learning RLHF DPO
20 Improving Reward Models with Synthetic Critiques 提出基于合成评论的奖励模型训练方法，提升数据效率与泛化能力。 reinforcement learning large language model instruction following
21 Towards Spoken Language Understanding via Multi-level Multi-grained Contrastive Learning 提出多层次多粒度对比学习框架MMCL，提升口语理解任务性能 contrastive learning distillation
22 SaySelf: Teaching LLMs to Express Confidence with Self-Reflective Rationales 提出SaySelf框架，提升LLM细粒度置信度表达能力并生成自反思性理由 reinforcement learning large language model ✅
23 Learning to Estimate System Specifications in Linear Temporal Logic using Transformers and Mamba 提出基于Transformer和Mamba的自回归模型，用于线性时序逻辑公式的系统规约挖掘。 Mamba
24 Self-Augmented Preference Optimization: Off-Policy Paradigms for Language Model Alignment 提出自增强偏好优化（SAPO），无需配对数据对语言模型进行对齐。 policy learning DPO direct preference optimization ✅

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
25 UniBias: Unveiling and Mitigating LLM Bias through Internal Attention and FFN Manipulation 提出UniBias以揭示和缓解LLM偏见问题 manipulation large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页