cs.CL（2025-06-12）

📊 共 45 篇论文 | 🔗 8 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (41 🔗8) 支柱二：RL算法与架构 (RL & Architecture) (4)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (41 篇)

#题目一句话要点标签🔗⭐
1 Burn After Reading: Do Multimodal Large Language Models Truly Capture Order of Events in Image Sequences? 提出TempVS基准以评估多模态大语言模型的事件顺序理解能力 large language model multimodal ✅
2 FloorPlan-DeepSeek (FPDS): A multimodal approach to floorplan generation using vector-based next room prediction 提出FPDS以解决建筑平面图生成的迭代性问题 large language model multimodal
3 Accelerating Diffusion Large Language Models with SlowFast Sampling: The Three Golden Principles 提出SlowFast采样以解决扩散语言模型的效率问题 large language model
4 Large Language Models for Detection of Life-Threatening Texts 利用大型语言模型检测生命威胁文本 large language model
5 BioPars: A Pretrained Biomedical Large Language Model for Persian Biomedical Text Mining 提出BioPars以解决波斯语生物医学文本挖掘问题 large language model ✅
6 Do Language Models Have Bayesian Brains? Distinguishing Stochastic and Deterministic Decision Patterns within Large Language Models 提出区分语言模型决策模式的方法以解决贝叶斯推断问题 large language model
7 Beyond Single-User Dialogue: Assessing Multi-User Dialogue State Tracking Capabilities of Large Language Models 评估大型语言模型在多用户对话状态跟踪中的能力 large language model
8 Team QUST at SemEval-2025 Task 10: Evaluating Large Language Models in Multiclass Multi-label Classification of News Entity Framing 提出三阶段检索框架以优化新闻实体框架的多类多标签分类 large language model ✅
9 FormosanBench: Benchmarking Low-Resource Austronesian Languages in the Era of Large Language Models 提出FORMOSANBENCH以评估低资源南岛语言的LLM表现 large language model
10 Code Execution as Grounded Supervision for LLM Reasoning 提出基于代码执行的监督方法以提升LLM推理能力 large language model chain-of-thought
11 ReCUT: Balancing Reasoning Length and Accuracy in LLMs via Stepwise Trails and Preference Optimization 提出ReCUT以解决大语言模型推理长度与准确性平衡问题 large language model chain-of-thought ✅
12 Conversational Search: From Fundamentals to Frontiers in the LLM Era 提出对话搜索系统以满足复杂信息需求 large language model instruction following
13 AC/DC: LLM-based Audio Comprehension via Dialogue Continuation 提出基于对话延续的音频理解模型以解决指令跟随问题 large language model instruction following
14 AutoMind: Adaptive Knowledgeable Agent for Automated Data Science 提出AutoMind以解决数据科学自动化中的灵活性不足问题 large language model ✅
15 ChineseHarm-Bench: A Chinese Harmful Content Detection Benchmark 提出ChineseHarm-Bench以解决中文有害内容检测不足问题 large language model ✅
16 Decomposing MLP Activations into Interpretable Features via Semi-Nonnegative Matrix Factorization 提出半非负矩阵分解以解析MLP激活特征 large language model
17 Generalization or Hallucination? Understanding Out-of-Context Reasoning in Transformers 提出统一机制以理解变换器中的上下文推理现象 large language model
18 UCD: Unlearning in LLMs via Contrastive Decoding 提出对比解码方法以解决大语言模型中的信息遗忘问题 large language model
19 PREMISE: Scalable and Strategic Prompt Optimization for Efficient Mathematical Reasoning in Large Models 提出PREMISE以解决大型推理模型的冗余计算问题 chain-of-thought
20 ClusterUCB: Efficient Gradient-Based Data Selection for Targeted Fine-Tuning of LLMs 提出ClusterUCB以解决大语言模型微调中的数据选择问题 large language model
21 No Universal Prompt: Unifying Reasoning through Adaptive Prompting for Temporal Table Reasoning 提出SEAR框架以解决时间表推理中的适应性提示问题 large language model
22 Hybrid-NL2SVA: Integrating RAG and Finetuning for LLM-based NL2SVA 提出Hybrid-NL2SVA框架以解决NL2SVA自动化问题 large language model
23 Slimming Down LLMs Without Losing Their Minds 提出基于LoRA和QLoRA的高效微调方法以提升大语言模型性能 large language model
24 NeuralNexus at BEA 2025 Shared Task: Retrieval-Augmented Prompting for Mistake Identification in AI Tutors 提出基于检索增强提示的AI辅导员错误识别系统 large language model ✅
25 Reliable Reasoning Path: Distilling Effective Guidance for LLM Reasoning with Knowledge Graphs 提出RRP框架以解决LLM推理中的知识图谱路径提取问题 large language model
26 ClimateChat: Designing Data and Methods for Instruction Tuning LLMs to Answer Climate Change Queries 提出自动化方法构建气候变化指令数据以提升LLM性能 large language model
27 Reasoning Isn't Enough: Examining Truth-Bias and Sycophancy in LLMs 评估大型语言模型的真实性偏见与谄媚行为 large language model
28 The Biased Samaritan: LLM biases in Perceived Kindness 提出一种新方法评估大型语言模型的偏见问题 large language model
29 Do We Still Need Audio? Rethinking Speaker Diarization with a Text-Based Approach Using Multiple Prediction Models 提出文本基础的说话人分离方法以解决音频质量问题 multimodal
30 From Replication to Redesign: Exploring Pairwise Comparisons for LLM-Based Peer Review 提出基于LLM的成对比较机制以优化同行评审流程 large language model
31 Dynamic Epistemic Friction in Dialogue 提出动态认知摩擦模型以优化人机对话中的信念更新 large language model
32 Beyond Gold Standards: Epistemic Ensemble of LLM Judges for Formal Mathematical Reasoning 提出基于LLM评估的自动化数学推理评估方法 large language model
33 Mitigating Negative Interference in Multilingual Sequential Knowledge Editing through Null-Space Constraints 提出LangEdit以解决多语言知识编辑中的负干扰问题 large language model ✅
34 Assessing RAG and HyDE on 1B vs. 4B-Parameter Gemma LLMs for Personal Assistants Integretion 评估RAG与HyDE在Gemma LLMs中的应用以提升个人助手性能 large language model
35 One Tokenizer To Rule Them All: Emergent Language Plasticity via Multilingual Tokenizers 提出通用分词器以提升多语言模型的适应能力 large language model
36 TaxoAdapt: Aligning LLM-Based Multidimensional Taxonomy Construction to Evolving Research Corpora 提出TaxoAdapt以解决科学文献分类与检索的动态问题 large language model
37 Inferring Adjective Hypernyms with Language Models to Increase the Connectivity of Open English Wordnet 提出基于语言模型的形容词上位词推断方法以增强开放英语Wordnet的连接性 large language model
38 Spelling-out is not Straightforward: LLMs' Capability of Tokenization from Token to Characters 揭示LLMs在字符拼写中的复杂性与内部表征 large language model
39 Surface Fairness, Deep Bias: A Comparative Study of Bias in Language Models 研究语言模型中的偏见问题及其影响 large language model
40 Beyond the Battlefield: Framing Analysis of Media Coverage in Conflict Reporting 利用计算方法分析冲突报道中的框架偏见 large language model
41 "Check My Work?": Measuring Sycophancy in a Simulated Educational Context 研究用户建议对大型语言模型的影响以解决教育公平问题 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (4 篇)

#题目一句话要点标签🔗⭐
42 Magistral 提出Magistral以实现大规模强化学习模型的推理能力 reinforcement learning multimodal instruction following
43 Fast on the Easy, Deep on the Hard: Efficient Reasoning via Powered Length Penalty 提出Powered Length Penalty以提高大语言模型推理效率 reinforcement learning large language model chain-of-thought
44 PAG: Multi-Turn Reinforced LLM Self-Correction with Policy as Generative Verifier 提出PAG框架以解决LLM自我验证与修正问题 reinforcement learning large language model
45 Beyond Random Sampling: Efficient Language Model Pretraining via Curriculum Learning 通过课程学习提升语言模型预训练效率 curriculum learning

⬅️ 返回 cs.CL 首页 · 🏠 返回主页