cs.CL（2025-12-31）

📊 共 15 篇论文

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (13) 支柱二：RL算法与架构 (RL & Architecture) (2)

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

#题目一句话要点标签🔗⭐
1 PCEval: A Benchmark for Evaluating Physical Computing Capabilities of Large Language Models PCEval：用于评估大语言模型物理计算能力的首个自动化基准 large language model
2 Do Large Language Models Know What They Are Capable Of? 评估大语言模型自我认知能力：模型能否准确预测自身任务表现？ large language model
3 Large language models and the entropy of English 利用大型语言模型揭示英语文本中的长程结构与依赖关系 large language model
4 Compute-Accuracy Pareto Frontiers for Open-Source Reasoning Large Language Models 针对开源推理大语言模型，构建计算-精度帕累托前沿，优化工业应用选型。 large language model
5 Understanding and Steering the Cognitive Behaviors of Reasoning Models at Test-Time 提出CREST，通过干预注意力头引导LLM推理，提升效率和准确率。 large language model chain-of-thought
6 Adaptive Dependency-aware Prompt Optimization Framework for Multi-Step LLM Pipeline 提出ADOPT框架，自适应优化多步LLM流水线中的提示，提升复杂任务性能。 large language model
7 Speculative Decoding: Performance or Illusion? 系统性分析推测解码在生产级LLM推理引擎中的加速效果与瓶颈 large language model
8 Encyclo-K: Evaluating LLMs with Dynamically Composed Knowledge Statements 提出Encyclo-K，通过动态组合知识语句评估LLM的综合理解能力 large language model
9 Safe in the Future, Dangerous in the Past: Dissecting Temporal and Linguistic Vulnerabilities in LLMs 揭示大语言模型在语言和时间维度上的安全漏洞，提出不变对齐 large language model
10 RIMRULE: Improving Tool-Using Language Agents via MDL-Guided Rule Learning RIMRULE：通过MDL引导的规则学习提升工具使用语言Agent能力 large language model
11 Vibe Coding, Interface Flattening 分析“Vibe Coding”范式，揭示LLM驱动开发中界面扁平化与控制权转移的矛盾 large language model
12 MUSIC: MUlti-Step Instruction Contrast for Multi-Turn Reward Models 提出MUSIC：多步指令对比方法，提升多轮对话奖励模型性能 large language model
13 Quantum Visual Word Sense Disambiguation: Unraveling Ambiguities Through Quantum Inference Model 提出量子推理模型以解决视觉词义消歧问题 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (2 篇)

#题目一句话要点标签🔗⭐
14 Adaptive Constraint Propagation: Scaling Structured Inference for Large Language Models via Meta-Reinforcement Learning 提出MetaJuLS，通过元强化学习加速大语言模型中的结构化推理。 reinforcement learning large language model
15 Youtu-LLM: Unlocking the Native Agentic Potential for Lightweight Large Language Models 提出Youtu-LLM，一种轻量级且具备原生Agent能力的语言模型，性能超越同规模模型。 distillation large language model

⬅️ 返回 cs.CL 首页 · 🏠 返回主页