cs.CL（2026-08-18）

📊 共 15 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (14 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (14 篇)

#题目一句话要点标签🔗⭐
1 BEAR-Bench: A Bilingual Enterprise and Academic Reasoning Benchmark for Multimodal Models 提出BEAR-Bench以解决多模态模型在专业文档推理中的不足 large language model multimodal
2 Multi-turn Conversational AI from Text to Multimodal Interaction: Data, Models, Evaluation, and Open Challenges 提出多模态对话AI以解决持续交互中的记忆和上下文问题 multimodal ✅
3 Do Large Language Models Play Six Degrees of Separation? Measuring Topological Compression in Long-Context Manifolds 提出几何分析方法以揭示大语言模型的推理机制 large language model
4 Auditing Exposure to Harmful Content on TikTok using Multimodal Language Models: A Cross-National, Age-Stratified Study 利用多模态语言模型审计TikTok有害内容的跨国研究 multimodal
5 Effects of Answer Format Variation on Gender Bias in Large Language Models 探讨回答格式变化对大型语言模型性别偏见的影响 large language model
6 An Investigation of Translationese in the Generations of Multilingual Large Language Models 研究多语言大语言模型生成的翻译特征以识别翻译语现象 large language model
7 Reflex-Guard: A Low-Latency Guardrail for LLM Prompt Safety Using Dense Semantic Embeddings 提出Reflex-Guard以解决LLM提示安全性问题 large language model
8 Encoded but Not Actionable: Auditing the Decode-Generate-Steer Gap in Frozen LLMs for Geometric Constraints 审计冻结LLM中的解码生成控制差距以解决几何约束问题 large language model
9 Interpretable Humans, Alien LLMs: Expert Analysis of Latent Structures in Assessment Responses 提出人类可解释性分析框架以评估大型语言模型的潜在结构 large language model
10 PTXBench: Benchmark and Adapt LLMs for GPU Kernel Optimization with Architecture-specific PTX 提出PTXBench以优化GPU内核的LLM适配问题 large language model
11 Chain-of-Experience for Continual LLM Improvement 提出Chain-of-Experience以提升大语言模型的持续学习能力 large language model
12 Whether LLMs Can Navigate Beliefs and Facts Depends on How You Phrase It 提出基于表达方式的信念与事实处理方法以提升LLM性能 large language model ✅
13 ArborMem: Navigating Interaction States with Memory Forests 提出ArborMem以解决对话状态记忆管理问题 large language model
14 Decomposition Attacks Across Unlinkable Identities: Limits of Stateful Defenses for LLM Services 提出状态监控机制以应对分解攻击挑战 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (1 篇)

#题目一句话要点标签🔗⭐
15 Write, Execute, Refine: From Skill Followers to Skill Optimizers via Reinforcement Learning from Execution Feedback 提出WER框架以优化工具使用代理的技能生成 reinforcement learning

⬅️ 返回 cs.CL 首页 · 🏠 返回主页