视觉-语言-动作论文趋势 2023–2026

VLA 与端到端驾驶

统计范围：NeurIPS、ICLR、ICML、CVPR 四个会议，2023–2026 年，共 1,639 篇被归入该主题的论文。

主要结论

每年论文数

各会议论文数

按年份和会议统计

视觉-语言-动作论文数，按年份和会议
年份NeurIPSICLRICMLCVPR合计
2023883034105257
20241082962130329
20251086433140345
2026204113126265708

代表论文

按年份和会议均匀挑选，方便快速看到每一届的样貌

全部视觉-语言-动作论文

年份全部2023202420252026
会议全部NeurIPSICLRICMLCVPR

共 708 篇，第 13 / 15 页

  1. NeurIPS 2026Long-term Embodied Visual Tracking with Lightweight Vision-Language-Action Models
  2. NeurIPS 2026LookThere! Sparse Vision by Reinforced Selection
  3. NeurIPS 2026Lost on Campus: Evaluating Embodied Spatial Reasoning of Vision-Language Models in the Wild
  4. NeurIPS 2026Masked Diffusion Vision-Language Models for Temporal Action Localization
  5. NeurIPS 2026Measuring Coherence in Predictive Models
  6. NeurIPS 2026MHWA: Multi-timescale Hierarchical World-Action Model
  7. NeurIPS 2026Mirror, Mirror on the Wall: Can VLM Agents Tell Who They Are at All?
  8. NeurIPS 2026Mitigating Asymmetric Boundary Encroachment in Continual Learning of Vision-Language Models
  9. NeurIPS 2026MM-SCALE: Evaluating Evidence-Grounded Moral Judgment in Vision-Language Models
  10. NeurIPS 2026MMTA: Benchmarking Multimodal Temporal Analysis with Time Series, Text, and Vision
  11. NeurIPS 2026Multi-Armed Bandits With Best-Action Queries
  12. NeurIPS 2026Near-Optimal Learning in Parametric Bandits with Action-Dependent Coarsened Feedback
  13. NeurIPS 2026NestedVLA: Learning to Consolidate and Generate Skills for Vision-Language-Action Model
  14. NeurIPS 2026Neural Garbage Collection: Learning to Forget while Learning to Reason
  15. NeurIPS 2026No-Regret Caching with Delayed Hits
  16. NeurIPS 2026Not Every Image Teaches Vision: Visual-Necessity-Gated Continual Learning for Multimodal Large Language Models
  17. NeurIPS 2026NoTVLA: Semantics-Preserving Robot Adaptation via Narrative Action Interfaces
  18. NeurIPS 2026NS-VLA: Towards Neuro-Symbolic Vision-Language-Action Models
  19. NeurIPS 2026Object Hallucination Mitigation in Large Vision-Language Models via Self-Vision Dual Masking and Uncertainty-Triggered Assembly
  20. NeurIPS 2026OGPO: Offline Goal-conditioned Policy Optimization for Recoverable Vision-Language-Action Models
  21. NeurIPS 2026OmniGF: A Dual-Branch Vision-Language Framework for Unified Gaze Following
  22. NeurIPS 2026On the Nature of Attention Sink that Shapes Decoding Strategy in Omni-LLMs
  23. NeurIPS 2026OT-Robust3DVLA: A Wasserstein Barycenter is the Right Inductive Bias for Robust Multi-View 3D Vision-Language-Action Policies
  24. NeurIPS 2026P$^3$-VLM: A Point-based Alternative for Grounded 3D Vision-Language Models
  25. NeurIPS 2026PAPO-VLA: Planning-Aware Policy Optimization for Vision-Language-Action Models
  26. NeurIPS 2026Patch Rebirth: Fast and Transferable Model Inversion of Vision Transformers
  27. NeurIPS 2026Pathways of Visual Information Flow in Vision-Language Models
  28. NeurIPS 2026Perception for Action in Latent World Models
  29. NeurIPS 2026Phase-Adaptive Fusion: Spatio-Temporal Modulation for VLA Models
  30. NeurIPS 2026Point Tracking Improves World Action Models
  31. NeurIPS 2026Poison-then-Hide: Finetuning-Activated Backdoor Attack on Pretrained Vision Encoders
  32. NeurIPS 2026PolyVision: Conditional Visual Scaling Via Dynamic Expert Routing For Vision-Centric MLLMs
  33. NeurIPS 2026PoseBridge: Bridging the Skeletonization Gap for Zero-Shot Skeleton-Based Action Recognition
  34. NeurIPS 2026Post-Training Quantization with Gradient-Projected Fisher Approximation for Vision Transformers
  35. NeurIPS 2026PriorVLA: Prior-Preserving Adaptation for Vision-Language-Action Models
  36. NeurIPS 2026PRISM: Primitive Routing via In-context Skill Mixing for Lifelong VLA
  37. NeurIPS 2026ProCLIP: Progressive Vision-Language Alignment via LLM-based Embedder
  38. NeurIPS 2026Profit Maximization in Bilateral Trade against a Smooth Adversary
  39. NeurIPS 2026Progressive Pseudo-label Self-balancing Towards Unsupervised Vision-Language Models Adaptation
  40. NeurIPS 2026Pure Exploration Beyond Reward Feedback: The Role of Post-Action Context
  41. NeurIPS 2026RadarFlowPose: Vision-Inspired Coarse-to-Fine Skeleton Refinement with Flow Matching
  42. NeurIPS 2026Realtime-VLA FLASH: Speculative Inference Framework for Diffusion-based VLAs
  43. NeurIPS 2026Remembering What Matters: From Markovian to Subtask-Causal Memory in VLA Policies
  44. NeurIPS 2026RePO-VLA: Recovery-Driven Policy Optimization for Vision-Language-Action Models
  45. NeurIPS 2026Residual Expertise Is Not Decision Value
  46. NeurIPS 2026Rethink Action Chunking in VLA Through Human Motor Control
  47. NeurIPS 2026Rethinking Causal Action Tokenization with Conditional Annealing in Flow Matching
  48. NeurIPS 2026Retrieval Heads Meet Vision: Uncovering How VLMs Locate and Extract Visual Information
  49. NeurIPS 2026Risk-Aware Action Repetition via Expected Skip Evaluation
  50. NeurIPS 2026ROCKET: Residual-Oriented Multi-Layer Alignment for Spatially-Aware Vision-Language-Action Models

在 Claude、Cursor 里直接问这个趋势

接上 Paperlist MCP 后，写综述、挑方向或准备投稿时，不用离开对话框就能拿到这页的数字：每年各会议的论文数、增长倍数、涨得最快的子方向和代表论文，AI 会按你的问题整理。

可以直接复制这句话问 AI

用 paperlist 查视觉-语言-动作在 CVPR 2023–2026 的论文趋势，哪一年涨得最快？

AI 会调用的工具：get_topic_stats {"topic": "vision-language-action"}

两分钟接入 Paperlist MCP（Claude Code、Cursor、Codex、ChatGPT）

常见问题

2023–2026 年四大会议一共有多少篇视觉-语言-动作论文？
共 1,639 篇，按年份分别是 2023 年 257 篇、2024 年 329 篇、2025 年 345 篇、2026 年 708 篇。
NeurIPS 2026 有多少篇视觉-语言-动作论文？
NeurIPS 2026 有 204 篇视觉-语言-动作论文，上一届是 108 篇。
ICLR 2026 有多少篇视觉-语言-动作论文？
ICLR 2026 有 113 篇视觉-语言-动作论文，上一届是 64 篇。
ICML 2026 有多少篇视觉-语言-动作论文？
ICML 2026 有 126 篇视觉-语言-动作论文，上一届是 33 篇。
CVPR 2026 有多少篇视觉-语言-动作论文？
CVPR 2026 有 265 篇视觉-语言-动作论文，上一届是 140 篇。
哪个会议的视觉-语言-动作论文最多？
CVPR，2023–2026 年共 640 篇。

「机器人与具身」下的其他主题