DAILY INTELLIGENCE · AI APP + VLA
研究日報研究日报
AI App 精選與 VLA 論文評級的每日合輯AI App 精选与 VLA 论文评级的每日合辑。 下方匯總近 7 天的信號總覽，歸檔索引可按領域過濾下方汇总近 7 天的信号总览，归档索引可按领域过滤。
近 7 日匯總近 7 日汇总 04-14 → 04-24
33 AI精選AI精选
157 VLA論文VLA论文
⚡ 4 突破
🔧 63 工具/技術工具/技术
📖 90 背景/觀點背景/观点
⚡ 本週最強信號本周最强信号
近 3 天內容近 3 天内容 04-22 → 04-24
2026-04-24 最新 44 條条
小米 MiMo-V2.5 发布：罗福莉领衔，对标 Claude Opus 4.6 / GPT-5.4 AI OpenAI Codex 推出 Automations：定时/触发式自动执行任务 AI OpenAI Codex 上线 Plugins & Skills 体系：连接外部工具 + 可复用工作流 AI LiteParse for the Web：纯浏览器端 PDF 空间文本解析 AI GitHub Trending：huggingface/ml-intern（开源ML工程师）和 claude-context（代码搜索MCP） AI Object-centric task representation and transfer using diffused orientation fields VLA From autonomy to alliance: Robotic foundation models must learn with us, not just for us VLA Boston Dynamics and Google DeepMind Teach Spot to Reason VLA JoyAI-RA 0.1: A Foundation Model for Robotic Autonomy VLA LLM-Guided Safety Agent for Edge Robotics with an ISO-Compliant Perception-Compute-Control Architecture VLA Cortex 2.0: Grounding World Models in Real-World Industrial Deployment VLA A Vision-Language-Action Model for Adaptive Ultrasound-Guided Needle Insertion and Needle Tracking VLA Bimanual Robot Manipulation via Multi-Agent In-Context Learning VLA Temporal Difference Calibration in Sequential Tasks: Application to Vision-Language-Action Models VLA FingerEye: Continuous and Unified Vision-Tactile Sensing for Dexterous Manipulation VLA Visual-Tactile Peg-in-Hole Assembly Learning from Peg-out-of-Hole Disassembly VLA PokeVLA: Empowering Pocket-Sized Vision-Language-Action Model with Comprehensive World Knowledge Guidance VLA Rodrigues Network for Learning Robot Actions VLA MATT-Diff: Multimodal Active Target Tracking by Diffusion Policy VLA OmniUMI: Towards Physically Grounded Robot Learning via Human-Aligned Multimodal Interaction VLA Evolvable Embodied Agent for Robotic Manipulation via Long Short-Term Reflection and Optimization VLA HELM: Harness-Enhanced Long-horizon Memory for Vision-Language-Action Manipulation VLA Gated Memory Policy VLA RoboWM-Bench: A Benchmark for Evaluating World Models in Robotic Manipulation VLA VLA Foundry: A Unified Framework for Training Vision-Language-Action Models VLA FASTER: Value-Guided Sampling for Fast RL VLA UniT: Toward a Unified Physical Language for Human-to-Humanoid Policy Learning and World Modeling VLA PhysMem: Scaling Test-time Physical Memory for Robot Manipulation VLA ExpertGen: Scalable Sim-to-Real Expert Policy Learning from Imperfect Behavior Priors VLA ARM: Advantage Reward Modeling for Long-Horizon Manipulation VLA If you're waiting for a sign... that might not be it! Mitigating Trust Boundary Confusion from Visual Injections on Vision-Language Agentic Systems VLA EmbodiedMidtrain: Bridging the Gap between Vision-Language Models and Vision-Language-Action Models via Mid-training VLA DeVI: Physics-based Dexterous Human-Object Interaction via Synthetic Video Imitation VLA Closed-loop tactile-visual interactivity via chip-free luminescent fibers enabled by capacitive coupling VLA A time-stamping tactile sensor enabled by pseudoconductive interface design at dielectric heterojunctions VLA ETac: A Lightweight and Efficient Tactile Simulation Framework for Learning Dexterous Manipulation VLA VTouch++: A Multimodal Dataset with Vision-Based Tactile Enhancement for Bimanual Manipulation VLA CubeDAgger: Interactive Imitation Learning for Dynamic Systems with Efficient yet Low-risk Interaction VLA CARLA-Air: Fly Drones Inside a CARLA World -- A Unified Infrastructure for Air-Ground Embodied Intelligence VLA Mask World Model: Predicting What Matters for Robust Robot Policy Learning VLA Curiosity-Critic: Cumulative Prediction Error Improvement as a Tractable Intrinsic Reward for World Model Training VLA Safety-Critical Contextual Control via Online Riemannian Optimization with World Models VLA Cloning Deterministic Worlds: The Critical Role of Latent Geometry in Long-Horizon World Models VLA X-Cache: Cross-Chunk Block Caching for Few-Step Autoregressive World Models Inference VLA
2026-04-23 昨日 24 條条
ChatGPT Images 2.0 发布：首个「会思考」的图像模型 AI Firefox 150 用 Claude Mythos 发现 271 个漏洞 AI OpenAI 推出 ChatGPT for Clinicians，对认证医师免费 AI AWS SageMaker AI 推出 GenAI 推理优化推荐 AI FASTER: Value-Guided Sampling for Fast RL VLA Gated Memory Policy VLA RoboWM-Bench: A Benchmark for Evaluating World Models in Robotic Manipulation VLA Wrench-Aware Admittance Control for Unknown-Payload Manipulation VLA VLA Foundry: A Unified Framework for Training Vision-Language-Action Models VLA UniT: Toward a Unified Physical Language for Human-to-Humanoid Policy Learning and World Modeling VLA PhysMem: Scaling Test-time Physical Memory for Robot Manipulation VLA ExpertGen: Scalable Sim-to-Real Expert Policy Learning from Imperfect Behavior Priors VLA ARM: Advantage Reward Modeling for Long-Horizon Manipulation VLA Drift-Based Policy Optimization: Native One-Step Policy Learning for Online Robot Control VLA HELM: Harness-Enhanced Long-horizon Memory for Vision-Language-Action Manipulation VLA LAMP: Lift Image-Editing as General 3D Priors for Open-world Manipulation VLA MultiWorld: Scalable Multi-Agent Multi-View Video World Models VLA Closed-loop tactile-visual interactivity via chip-free luminescent fibers enabled by capacitive coupling VLA A time-stamping tactile sensor enabled by pseudoconductive interface design at dielectric heterojunctions VLA Mask World Model: Predicting What Matters for Robust Robot Policy Learning VLA Flow-Opt: Scalable Centralized Multi-Robot Trajectory Optimization with Flow Matching and Differentiable Optimization VLA Curiosity-Critic: Cumulative Prediction Error Improvement as a Tractable Intrinsic Reward for World Model Training VLA Safety-Critical Contextual Control via Online Riemannian Optimization with World Models VLA Cloning Deterministic Worlds: The Critical Role of Latent Geometry in Long-Horizon World Models VLA
2026-04-22 2日前 35 條条
SpaceX 洽以 600 亿美元收购 Cursor AI Kimi K2.6 正式开源：SWE-Bench Pro 58.6%、300 Agent 集群 AI Claude Code 从 Pro 套餐中移除 AI Vercel OAuth 供应链攻击：Context.ai 泄露成突破口 AI CrabTrap：Brex 开源 LLM-as-a-judge Agent 安全代理 AI nateherkai/token-dashboard：Claude Code token 消耗可视化 AI Demonstrate once, execute on many: Kinematic intelligence for cross-robot skill transfer VLA A careful examination of large behavior models for multitask dexterous manipulation VLA XEmbodied: A Foundation Model with Enhanced Geometric and Physical Cues for Large-Scale Embodied Environments VLA Test-Time Perturbation Learning with Delayed Feedback for Vision-Language-Action Models VLA Disentangled Robot Learning via Separate Forward and Inverse Dynamics Pretraining VLA ReconVLA: An Uncertainty-Guided and Failure-Aware Vision-Language-Action Framework for Robotic Control VLA Rewind-IL: Online Failure Detection and State Respawning for Imitation Learning VLA LongBench: Evaluating Robotic Manipulation Policies on Real-World Long-Horizon Tasks VLA Chain Of Interaction Benchmark (COIN): When Reasoning meets Embodied Interaction VLA GaLa: Hypergraph-Guided Visual Language Models for Procedural Planning VLA FLASH: Fast Learning via GPU-Accelerated Simulation for High-Fidelity Deformable Manipulation in Minutes VLA OmniVLA-RL: A Vision-Language-Action Model with Spatial Understanding and Online RL VLA AnchorRefine: Synergy-Manipulation Based on Trajectory Anchor and Residual Refinement for Vision-Language-Action Models VLA ReFineVLA: Multimodal Reasoning-Aware Generalist Robotic Policies via Teacher-Guided Fine-Tuning VLA OFlow: Injecting Object-Aware Temporal Flow Matching for Robust Robotic Manipulation VLA ST-$\pi$: Structured SpatioTemporal VLA for Robotic Manipulation VLA StableIDM: Stabilizing Inverse Dynamics Model against Manipulator Truncation via Spatio-Temporal Refinement VLA Unmasking the Illusion of Embodied Reasoning in Vision-Language-Action Models VLA COFFAIL: A Dataset of Successful and Anomalous Robot Skill Executions in the Context of Coffee Preparation VLA Can Explicit Physical Feasibility Benefit VLA Learning? An Empirical Study VLA SafeVLA: Towards Safety Alignment of Vision-Language-Action Model via Constrained Learning VLA On the Importance of Tactile Sensing for Imitation Learning: A Case Study on Robotic Match Lighting VLA UniDomain: Pretraining a Unified PDDL Domain from Real-World Demonstrations for Generalizable Robot Task Planning VLA Contact-Rich Robotic Assembly in Construction via Diffusion Policy Learning VLA Stable Language Guidance for Vision-Language-Action Models VLA ROBOGATE: Adaptive Failure Discovery for Safe Robot Policy Deployment via Two-Stage Boundary-Focused Sampling VLA World-Value-Action Model: Implicit Planning for Vision-Language-Action Systems VLA InternScenes: A Large-scale Simulatable Indoor Scene Dataset with Realistic Layouts VLA DeepThinkVLA: Enhancing Reasoning Capability of Vision-Language-Action Models VLA
歸檔索引归档索引
73 期— 2026 年 4 月 —
AI 最新
5 篇
VLA 最新
39 篇
AI 昨日
4 篇
VLA 昨日
20 篇
AI 2日前
6 篇
VLA 2日前
29 篇
AI
6 篇
AI
3 篇
AI
3 篇
AI
6 篇
VLA
15 篇
AI
6 篇
VLA
20 篇
AI
8 篇
VLA
17 篇
AI
6 篇
AI
5 篇
VLA
17 篇
AI
2 篇
AI
3 篇
VLA
2 篇
AI
3 篇
VLA
28 篇
AI
6 篇
VLA
20 篇
AI
7 篇
VLA
26 篇
AI
6 篇
VLA
37 篇
AI
5 篇
VLA
15 篇
AI
5 篇
AI
6 篇
AI
6 篇
VLA
25 篇
AI
5 篇
VLA
10 篇
AI
4 篇
VLA
8 篇
AI
7 篇
VLA
28 篇
— 2026 年 3 月 —
AI
5 篇
VLA
17 篇
AI
6 篇
AI
6 篇
AI
6 篇
VLA
35 篇
VLA
24 篇
AI
8 篇
VLA
22 篇
AI
8 篇
VLA
11 篇
VLA
10 篇
VLA
1 篇
VLA
13 篇
VLA
12 篇
VLA
3 篇
VLA
30 篇
VLA
3 篇
VLA
11 篇
VLA
11 篇
VLA
29 篇
VLA
16 篇
VLA
1 篇
VLA
9 篇
VLA
9 篇
VLA
24 篇
VLA
20 篇
VLA
1 篇
VLA
1 篇
— 2026 年 2 月 —
VLA
3 篇
VLA
25 篇
VLA
26 篇