cs.CL（2025-09-17）

📊 共 38 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (32 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (4) 支柱八：物理动画 (Physics-based Animation) (1) 支柱七：动作重定向 (Motion Retargeting) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (32 篇)

#题目一句话要点标签🔗⭐
1 Early Stopping Chain-of-thoughts in Large Language Models 提出ES-CoT，通过提前停止CoT生成降低大语言模型推理成本 large language model chain-of-thought
2 Simulating a Bias Mitigation Scenario in Large Language Models 构建模拟框架，评估缓解大型语言模型偏见的策略 large language model
3 Annotating Training Data for Conditional Semantic Textual Similarity Measurement using Large Language Models 利用大型语言模型重新标注条件语义文本相似度数据集，提升模型性能。 large language model ✅
4 Do Large Language Models Understand Word Senses? 评估大型语言模型对词义理解能力，并验证其在词义消歧任务中的有效性 large language model
5 Large Language Models Discriminate Against Speakers of German Dialects 大型语言模型对德语方言使用者存在歧视性偏见 large language model
6 How Can Quantum Deep Learning Improve Large Language Models? 探索量子深度学习在提升大型语言模型适应性方面的潜力 large language model
7 DSCC-HS: A Dynamic Self-Reinforcing Framework for Hallucination Suppression in Large Language Models 提出DSCC-HS框架以主动抑制大型语言模型的幻觉现象 large language model
8 Can Large Language Models Robustly Perform Natural Language Inference for Japanese Comparatives? 构建日语比较句NLI数据集，评估大语言模型在此任务上的鲁棒性 large language model
9 AssoCiAm: A Benchmark for Evaluating Association Thinking while Circumventing Ambiguity 提出AssoCiAm基准，通过混合计算方法评估多模态大语言模型的联想思维能力，并规避歧义性。 large language model multimodal
10 Enhancing Time Awareness in Generative Recommendation 提出GRUT模型，通过时间感知提升生成式推荐效果 large language model TAMP ✅
11 Integrating Text and Time-Series into (Large) Language Models to Predict Medical Outcomes 利用DSPy优化提示，将文本和时间序列融入LLM以预测医疗结果 large language model multimodal
12 Estimating Semantic Alphabet Size for LLM Uncertainty Quantification 提出改进的语义字母表大小估计器，提升LLM不确定性量化的准确性和可解释性 large language model
13 Ticket-Bench: A Kickoff for Multilingual and Regionalized Agent Evaluation Ticket-Bench：多语言区域化Agent评估基准，提升真实场景任务性能 large language model
14 Correct-Detect: Balancing Performance and Ambiguity Through the Lens of Coreference Resolution in LLMs 揭示LLM在共指消解中性能与歧义检测的权衡：Correct-Detect困境 large language model
15 Causal-Counterfactual RAG: The Integration of Causal-Counterfactual Reasoning into RAG 提出因果-反事实RAG，将因果推理融入RAG以提升知识密集型任务性能 large language model
16 Adding LLMs to the psycholinguistic norming toolbox: A practical guide to getting the most out of human ratings 提出一种利用大型语言模型增强心理语言学规范数据集的方法 large language model
17 Apertus: Democratizing Open and Compliant LLMs for Global Language Environments Apertus：构建开放、合规且支持全球语言环境的大语言模型 large language model
18 ShinkaEvolve: Towards Open-Ended And Sample-Efficient Program Evolution ShinkaEvolve：提出一种高效、开源的程序进化框架，用于解决科学发现中的样本效率问题。 large language model
19 Enhancing Multi-Agent Debate System Performance via Confidence Expression 提出ConfMAD框架，通过置信度表达提升多智能体辩论系统性能 large language model
20 Hala Technical Report: Building Arabic-Centric Instruction & Translation Models at Scale 提出Hala模型以提升阿拉伯语指令与翻译的质量 instruction following
21 Do LLMs Align Human Values Regarding Social Biases? Judging and Explaining Social Biases with LLMs 评估大语言模型在社会偏见场景下的人类价值观对齐程度 large language model
22 Characterizing Knowledge Graph Tasks in LLM Benchmarks Using Cognitive Complexity Frameworks 利用认知复杂性框架表征LLM基准测试中的知识图谱任务 large language model
23 Exploring Data and Parameter Efficient Strategies for Arabic Dialect Identifications 探索数据与参数高效的阿拉伯语方言识别策略 large language model
24 Thinking in a Crowd: How Auxiliary Information Shapes LLM Reasoning 研究辅助信息对LLM推理的影响：有害信息会显著降低LLM的推理能力 large language model ✅
25 Implementing a Logical Inference System for Japanese Comparatives 提出ccg-jcomp，用于日语比较句自然语言推理的逻辑推理系统。 large language model
26 DSPC: Dual-Stage Progressive Compression Framework for Efficient Long-Context Reasoning 提出DSPC双阶段渐进压缩框架，无需训练即可高效压缩长文本上下文，提升LLM推理效率。 large language model
27 Improving Context Fidelity via Native Retrieval-Augmented Reasoning 提出CARE框架，通过原生检索增强推理提升LLM上下文忠实度 large language model
28 A Simple and Efficient Jailbreak Method Exploiting LLMs' Helpfulness HILL：一种利用LLM助人为乐特性进行越狱的简单高效方法 large language model
29 CL$^2$GEC: A Multi-Discipline Benchmark for Continual Learning in Chinese Literature Grammatical Error Correction 提出CL$^2$GEC基准，用于评估中文语法纠错系统在多领域持续学习中的性能。 large language model
30 Sparse Neurons Carry Strong Signals of Question Ambiguity in LLMs 发现LLM中编码问题歧义的稀疏神经元，实现歧义检测与行为控制 large language model
31 ZERA: Zero-init Instruction Evolving Refinement Agent -- From Zero Instructions to Structured Prompts via Principle-based Optimization ZERA：零初始化指令演化优化Agent，通过基于原则的优化从零指令生成结构化提示 large language model ✅
32 Latent Traits and Cross-Task Transfer: Deconstructing Dataset Interactions in LLM Fine-tuning 通过潜在特征和跨任务迁移，解构LLM微调中的数据集交互 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🔗⭐
33 Process-Supervised Reinforcement Learning for Interactive Multimodal Tool-Use Agents 提出Turn-level Adjudicated RL，解决交互式多模态工具使用Agent的训练难题 reinforcement learning large language model foundation model
34 Teaching According to Talents! Instruction Tuning LLMs with Competence-Aware Curriculum Learning 提出CAMPUS框架，通过能力感知课程学习提升指令调优大语言模型性能 curriculum learning large language model
35 SSL-SSAW: Self-Supervised Learning with Sigmoid Self-Attention Weighting for Question-Based Sign Language Translation 提出基于问题的手语翻译任务及自监督学习融合方法SSL-SSAW contrastive learning multimodal
36 Noise Supervised Contrastive Learning and Feature-Perturbed for Anomalous Sound Detection 提出噪声监督对比学习与特征扰动方法，提升异常声音检测精度并减少误报。 contrastive learning

🔬 支柱八：物理动画 (Physics-based Animation) (1 篇)

#题目一句话要点标签🔗⭐
37 Video2Roleplay: A Multimodal Dataset and Framework for Video-Guided Role-playing Agents 提出Video2Roleplay框架，通过视频信息增强角色扮演Agent的动态感知能力 interactive character multimodal

🔬 支柱七：动作重定向 (Motion Retargeting) (1 篇)

#题目一句话要点标签🔗⭐
38 Geometric Uncertainty for Detecting and Correcting Hallucinations in LLMs 提出基于几何不确定性的方法，用于检测和纠正大语言模型中的幻觉问题。 spatial relationship large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页