cs.CL（2026-02-25）

📊 共 23 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) (16 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (6) 支柱六：视频提取与匹配 (Video Extraction) (1)

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

#题目一句话要点标签🔗⭐
1 MEDSYN: Benchmarking Multi-EviDence SYNthesis in Complex Clinical Cases for Multimodal Large Language Models MEDSYN：多模态大语言模型在复杂临床病例中多证据综合的基准测试 large language model multimodal
2 FewMMBench: A Benchmark for Multimodal Few-Shot Learning 提出FewMMBench，用于评估多模态大语言模型的小样本学习能力 large language model multimodal chain-of-thought ✅
3 Understanding Artificial Theory of Mind: Perturbed Tasks and Reasoning in Large Language Models 通过扰动任务和思维链推理评估大型语言模型中的心理理论能力 large language model chain-of-thought
4 Scalable Multilingual Multimodal Machine Translation with Speech-Text Fusion 提出基于语音-文本融合的可扩展多语言多模态机器翻译框架，显著提升翻译质量。 large language model multimodal ✅
5 Dynamic Personality Adaptation in Large Language Models via State Machines 提出基于状态机的大语言模型动态人格适应框架，用于改善人机交互 large language model
6 Sparsity Induction for Accurate Post-Training Pruning of Large Language Models 提出稀疏性诱导方法，提升大语言模型后训练剪枝的准确性 large language model
7 IndicIFEval: A Benchmark for Verifiable Instruction-Following Evaluation in 14 Indic Languages IndicIFEval：面向14种印度语言的可验证指令遵循评估基准 instruction following ✅
8 Large Language Models are Algorithmically Blind 揭示大语言模型在算法推理上的局限性：算法盲区 large language model
9 Small Wins Big: Comparing Large Language Models and Domain Fine-Tuned Models for Sarcasm Detection in Code-Mixed Hinglish Text 在Hinglish混合语境下，领域微调DistilBERT胜过大型语言模型，用于反讽检测 large language model
10 Personalized Graph-Empowered Large Language Model for Proactive Information Access 提出基于个性化图谱增强的大语言模型，用于主动信息访问以辅助记忆 large language model
11 Evaluating the Usage of African-American Vernacular English in Large Language Models 评估大型语言模型中非裔美国人白话英语的使用情况，揭示其不足与刻板印象。 large language model
12 When More Is Less: A Systematic Analysis of Spatial and Commonsense Information for Visual Spatial Reasoning 针对视觉空间推理，分析空间信息和常识信息注入策略的有效性 multimodal chain-of-thought
13 Recovered in Translation: Efficient Pipeline for Automated Translation of Benchmarks and Datasets 提出T-RANK框架，提升多语言LLM评测基准翻译质量，解决语义漂移问题。 large language model
14 Confidence-Driven Multi-Scale Model Selection for Cost-Efficient Inference 提出置信度驱动的多尺度模型选择策略，实现成本效益高的LLM推理 large language model
15 CxMP: A Linguistic Minimal-Pair Benchmark for Evaluating Constructional Understanding in Language Models 提出CxMP基准以评估语言模型的构式理解能力 large language model
16 Improving Implicit Discourse Relation Recognition with Natural Language Explanations from LLMs 利用大语言模型解释增强隐式篇章关系识别 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (6 篇)

#题目一句话要点标签🔗⭐
17 RuCL: Stratified Rubric-Based Curriculum Learning for Multimodal Large Language Model Reasoning 提出基于分层评分准则课程学习的RuCL框架，提升多模态大语言模型的推理能力。 reinforcement learning curriculum learning reward design
18 D-COT: Disciplined Chain-of-Thought Learning for Efficient Reasoning in Small Language Models 提出D-CoT，通过控制标签约束思维链，提升小模型推理效率与准确率。 distillation large language model chain-of-thought
19 Explore-on-Graph: Incentivizing Autonomous Exploration of Large Language Models on Knowledge Graphs with Path-refined Reward Modeling 提出Explore-on-Graph框架，激励LLM在知识图谱上自主探索推理路径 reinforcement learning large language model
20 DWA-KD: Dual-Space Weighting and Time-Warped Alignment for Cross-Tokenizer Knowledge Distillation DWA-KD：双空间权重与时间扭曲对齐的跨分词器知识蒸馏 distillation large language model
21 Evaluating the relationship between regularity and learnability in recursive numeral systems using Reinforcement Learning 利用强化学习评估递归数字系统中规则性与可学习性的关系 reinforcement learning
22 RADAR: Reasoning as Discrimination with Aligned Representations for LLM-based Knowledge Graph Reasoning 提出RADAR，通过对齐表示的判别推理增强LLM在知识图谱推理中的能力 reinforcement learning large language model

🔬 支柱六：视频提取与匹配 (Video Extraction) (1 篇)

#题目一句话要点标签🔗⭐
23 MixSarc: A Bangla-English Code-Mixed Corpus for Implicit Meaning Identification 提出MixSarc：孟加拉-英语混合语料库，用于隐式含义识别。 HuMoR large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页