cs.AI（2026-08-05）

📊 共 21 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) (16 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (5)

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

#题目一句话要点标签🔗⭐
1 When Prompts Become Pixels: Prompt-Region Grounding for Multimodal Reasoning 提出提示区域对齐以解决多模态推理中的语义通道差距问题 large language model multimodal
2 MarsCast: Transfer Learning of AI Weather Foundation Models to Planetary Atmospheres 提出MarsCast以解决地球天气模型在火星大气预测中的适用性问题 foundation model
3 Chain-of-Thought Monitoring Can Be Unreliable in Implicit-Influence Settings 提出隐性影响下的链式思维监控评估基准 chain-of-thought ✅
4 Short-term load forecasting under EU-AI Act Requirements in Safety-Critical Environments: Results from a 41-day live challenge on the aggregated German transmission-grid load 提出基于EU-AI法案的短期负荷预测方法以应对安全关键环境挑战 foundation model
5 Hardware Design and Security in the Era of Chiplets and LLMs 提出统一分析以应对芯片和大语言模型时代的硬件安全挑战 large language model
6 The Effect of Perceived Race and Gender on Police Language Use: Experimental Evidence from VR Simulations 探讨种族与性别对警察语言使用的影响 large language model
7 Gradient Immunity: Null-Space Resistance to Malicious Fine-Tuning 提出梯度免疫机制以应对恶意微调问题 large language model ✅
8 From Score Matrices to Football-Aware Match-State Simulation: An Auditable LLM Harness for Exact-Score Reranking 提出审计化LLM框架以改进足球比分预测 large language model
9 SciCode-Verified: How Benchmark Defects Underestimated the Scientific-Coding Ability of Language Models 提出SciCode-Verified以解决科学编码能力评估的缺陷问题 instruction following
10 RepoProbe: Benchmarking Architecture-Aware Repository Comprehension with Checklists 提出RepoProbe以解决现有代码理解基准的不足问题 large language model
11 Fewer Tokens, Smaller Cache: Reward-Coordinated Efficient Reasoning 提出ReCo以解决长链推理中的缓存效率问题 chain-of-thought
12 Active-SWE: Benchmarking Coding Agents for Proactive Bug Fixing without Issue Reports 提出Active-SWE以解决缺乏问题报告的主动修复挑战 large language model
13 Leak-Resistant Unlearning: A New Benchmark for Evaluating Multi-Hop Reasoning Consistency and Recovery Robustness 提出Leak-Resistant Unlearning基准以评估多跳推理一致性与恢复鲁棒性 large language model
14 CARGO-VL: Counterfactual Arbitration with Risk-Constrained Group Optimization for Vision-Language Models 提出CARGO-VL以解决视觉语言模型中的反事实仲裁问题 multimodal
15 ExeCRE: Execution-Consistency Guided Reliability Estimation for Self-Correcting Code Generation 提出ExeCRE框架以解决自校正代码生成中的可靠性估计问题 large language model
16 Combating Knowledge Corruption in Agent Systems: A Byzantine-Tolerant Secure Collaborative RAG Framework 提出SecureCollaRAG以解决知识腐蚀问题 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (5 篇)

#题目一句话要点标签🔗⭐
17 WorldCycle: Self-Verifiable Reinforcement Learning for Long-Horizon Video World Models 提出WorldCycle以解决长视频世界模型的验证瓶颈问题 reinforcement learning world model world models
18 Trident : How to Break Deep Reinforcement Learning Cyber Defenses (Agentic) 提出Trident框架以增强深度强化学习网络防御能力 reinforcement learning deep reinforcement learning DRL
19 Calibrating Artificial Guilt: Neurally Grounded Reward Shaping for Prosocial Multi-Agent Reinforcement Learning 提出基于神经数据的人工内疚信号以优化多智能体强化学习 reinforcement learning reward shaping
20 Privileged, but Biased: How PI-Conditioned Teachers Break Self-Distillation 揭示PI条件教师如何破坏自蒸馏的有效性 reinforcement learning distillation privileged information
21 Joint UAV Flight and Opportunistic Routing under Reinforcement Learning for Delay-Tolerant Networks 提出联合无人机飞行与机会路由以解决延迟容忍网络中的性能问题 reinforcement learning PPO

⬅️ 返回 cs.AI 首页 · 🏠 返回主页