cs.CL（2026-03-11）

📊 共 19 篇论文 | 🔗 1 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (13 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (6)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (13 篇)

#题目一句话要点标签🔗⭐
1 Aligning Large Language Models with Searcher Preferences 提出SearchLLM，用于开放域生成式搜索，提升用户体验并保证安全性。 large language model
2 Large Language Models as Annotators for Machine Translation Quality Estimation 利用大语言模型生成标注数据，提升机器翻译质量评估模型性能 large language model
3 Word Recovery in Large Language Models Enables Character-Level Tokenization Robustness 通过词语恢复机制，大语言模型实现字符级分词的鲁棒性 large language model
4 Prism-$Δ$: Differential Subspace Steering for Prompt Highlighting in Large Language Models 提出PRISM-$Δ$，通过差异子空间引导实现大语言模型中的Prompt高亮。 large language model
5 Automatic End-to-End Data Integration using Large Language Models 提出基于大型语言模型的端到端自动数据集成方案，降低人工成本。 large language model
6 Speech Codec Probing from Semantic and Phonetic Perspectives 分析语音编码器语义与音素信息，揭示其对多模态LLM性能的影响 large language model multimodal
7 Mitigating Translationese Bias in Multilingual LLM-as-a-Judge via Disentangled Information Bottleneck DIBJudge：通过解耦信息瓶颈缓解多语言LLM评估中的翻译腔偏差 large language model
8 Is this Idea Novel? An Automated Benchmark for Judgment of Research Ideas 提出RINoBench，用于大规模自动化评估研究想法新颖性的基准测试。 large language model ✅
9 A Systematic Study of Pseudo-Relevance Feedback with LLMs 系统性研究LLM伪相关反馈，揭示反馈源与反馈模型对检索效果的影响 large language model
10 GLM-OCR Technical Report GLM-OCR：一种高效紧凑的多模态模型，用于现实文档理解任务。 multimodal
11 PivotAttack: Rethinking the Search Trajectory in Hard-Label Text Attacks via Pivot Words 提出PivotAttack，通过枢轴词重塑硬标签文本攻击的搜索轨迹 large language model
12 Making Bielik LLM Reason (Better): A Field Report 评估并提升波兰语LLM Bielik的推理能力 large language model
13 Disentangling Similarity and Relatedness in Topic Models 提出一种解耦主题模型中相似性和相关性的方法，用于更细粒度的模型评估和任务预测。 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🔗⭐
14 LLM2Vec-Gen: Generative Embeddings from Large Language Models LLM2Vec-Gen：利用大型语言模型生成式嵌入，提升文本表示能力。 contrastive learning distillation large language model
15 Large language models can disambiguate opioid slang on social media 利用大型语言模型消除社交媒体上阿片类药物俚语的歧义 predictive model large language model
16 VERI-DPO: Evidence-Aware Alignment for Clinical Summarization via Claim Verification and Direct Preference Optimization 提出VERI-DPO以解决临床摘要中的证据对齐问题 DPO direct preference optimization
17 From Images to Words: Efficient Cross-Modal Knowledge Distillation to Language Models from Black-box Teachers 提出ARMADA，通过高效跨模态知识蒸馏，从黑盒视觉-语言模型迁移知识到纯语言模型。 distillation multimodal
18 Learning to Negotiate: Multi-Agent Deliberation for Collective Value Alignment in LLMs 提出基于多智能体协商的LLM对齐框架，提升集体价值对齐与冲突解决能力 RLHF large language model
19 mAceReason-Math: A Dataset of High-Quality Multilingual Math Problems Ready For RLVR 提出mAceReason-Math，一个高质量多语言数学问题数据集，用于促进基于可验证奖励的强化学习研究。 reinforcement learning large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页