推理模型论文趋势 2023–2026

主要结论

• 推理模型论文从 2023 年的 470 篇增长到 2026 年的 1,799 篇，三年间扩大到约 3.8 倍。
• NeurIPS 是发推理模型论文最多的会议，2023–2026 年共 1,253 篇，占四个会议总量的 36%。
• 交叉方向里涨得最快的是强化学习：同时属于两个主题的论文从 2023 年的 9 篇增至 2026 年的 437 篇。
• 在「前沿系统」的 4 个主题里，它的 2026 年论文数排第 3；同组增长最快的AI 智能体是 12 倍，推理模型是 3.8 倍。

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

• NeurIPS 2026A Mechanistic Analysis of Looped Reasoning Language Models
• ICML 2026A Formal Comparison Between Chain of Thought and Latent Thought
• ICLR 2026A Simple "Motivation" Can Enhance Reinforcement Finetuning of Large Reasoning Models
• CVPR 2026Asking like Socrates Socrates helps VLMs understand remote sensing images
• NeurIPS 2025A Frustratingly Simple Yet Highly Effective Attack Baseline: Over 90% Success Rate Against the Strong Black-box Models of GPT-4.5/4o/o1
• ICML 2025A Reasoning-Based Approach to Cryptic Crossword Clue Solving
• ICLR 2025A Formal Framework for Understanding Length Generalization in Transformers
• CVPR 2025Argus: Vision-Centric Reasoning with Grounded Chain-of-Thought
• NeurIPS 2024Are Large Language Models Good Statisticians?
• ICML 2024Algorithm of Thoughts: Enhancing Exploration of Ideas in Large Language Models
• ICLR 2024Amortizing intractable inference in large language models
• CVPR 2024Generate Subgoal Images before Act: Unlocking the Chain-of-Thought Reasoning in Diffusion Model for Robot Manipulation with Multimodal Prompts
• NeurIPS 2023A benchmark of categorical encoders for binary classification
• ICML 2023Can We Scale Transformers to Predict Parameters of Diverse ImageNet Models?
• ICLR 2023Automatic Chain of Thought Prompting in Large Language Models
• CVPR 2023ActMAD: Activation Matching To Align Distributions for Test-Time-Training
• NeurIPS 2026A Solvable Model of Chain-of-Thought in In-Context Learning
• ICML 2026A Narrowing Geometry in Contaminated Reasoning
• ICLR 2026A State-Transition Framework for Efficient LLM Reasoning
• CVPR 2026Beyond Weak Supervision MLLMs-Guided Graded Knowledge Distillation for Unsupervised Camouflaged

全部推理模型论文

共 1,799 篇，第 9 / 36 页

401. ICLR 2026GeoGramBench: Benchmarking the Geometric Program Reasoning in Modern LLMs
402. ICLR 2026Get RICH or Die Scaling: Profitably Trading Inference Compute for Robustness
403. ICLR 2026GHOST: Hallucination-Inducing Image Generation for Multimodal LLMs
404. ICLR 2026GIQ: Benchmarking 3D Geometric Reasoning of Vision Foundation Models with Simulated and Real Polyhedra
405. ICLR 2026Global-Recent Semantic Reasoning on Dynamic Text-Attributed Graphs with Large Language Models
406. ICLR 2026Goedel-Prover-V2: Scaling Formal Theorem Proving with Scaffolded Data Synthesis and Self-Correction
407. ICLR 2026GoT-R1: Unleashing Reasoning Capability of Autoregressive Visual Generation with Reinforcement Learning
408. ICLR 2026GPG: A Simple and Strong Reinforcement Learning Baseline for Model Reasoning
409. ICLR 2026Grokking in LLM Pretraining? Monitor Memorization-to-Generalization without Test
410. ICLR 2026Grounding Generative Planners in Verifiable Logic: A Hybrid Architecture for Trustworthy Embodied AI
411. ICLR 2026GTR-Bench: Evaluating Geo-Temporal Reasoning in Vision-Language Models
412. ICLR 2026Guided Speculative Inference for Efficient Test-Time Alignment of LLMs
413. ICLR 2026HalluGuard: Demystifying Data-Driven and Reasoning-Driven Hallucinations in LLMs
414. ICLR 2026HardcoreLogic: Challenging Large Reasoning Models with Long-tail Logic Puzzle Games
415. ICLR 2026Harder Is Better: Boosting Mathematical Reasoning via Difficulty-Aware GRPO and Multi-Aspect Question Reformulation
416. ICLR 2026HARP: Hallucination Detection via Reasoning Subspace Projection
417. ICLR 2026HeurekaBench: A Benchmarking Framework for AI Co-scientist
418. ICLR 2026Hidden Patterns in Chain-of-Thought Reasoning
419. ICLR 2026HiPO: Self-Hint Policy Optimization for RLVR
420. ICLR 2026Hippoformer: Integrating Hippocampus-inspired Spatial Memory with Transformers
421. ICLR 2026How Dark Patterns Manipulate Web Agents
422. ICLR 2026How Far Are LLMs from Professional Poker Players? Revisiting Game-Theoretic Reasoning with Agentic Tool Use
423. ICLR 2026How Far Can Unsupervised RLVR Scale LLM Training?
424. ICLR 2026How reinforcement learning after next-token prediction facilitates learning
425. ICLR 2026How Well Does GPT-4o Understand Vision? Evaluating Multimodal Foundation Models on Standard Computer Vision Tasks
426. ICLR 2026HumanPCR: Probing MLLM Capabilities in Diverse Human-Centric Scenes
427. ICLR 2026Hybrid Deep Searcher: Scalable Parallel and Sequential Search Reasoning
428. ICLR 2026Hybrid Training for Vision-Language-Action Models
429. ICLR 2026IceCache: Memory-Efficient KV-cache Management for Long-Sequence LLMs
430. ICLR 2026ICPO: Provable and Practical In-Context Policy Optimization for Test-Time Scaling
431. ICLR 2026IGC-Net for conditional average potential outcome estimation over time
432. ICLR 2026Implicit Models: Expressive Power Scales with Test-Time Compute
433. ICLR 2026In-Context Algebra
434. ICLR 2026In-Place Test-Time Training
435. ICLR 2026Incentivizing LLM Reasoning via Reinforcement Learning with Functional Monte Carlo Tree Search
436. ICLR 2026Inferring the Invisible: Neuro-Symbolic Rule Discovery for Missing Value Imputation
437. ICLR 2026InfoDet: A Dataset for Infographic Element Detection
438. ICLR 2026InftyThink: Breaking the Length Limits of Long-Context Reasoning in Large Language Models
439. ICLR 2026Interleaving Reasoning for Better Text-to-Image Generation
440. ICLR 2026Internal Planning in Language Models: Characterizing Horizon and Branch Awareness
441. ICLR 2026InternSpatial: A Comprehensive Dataset for Spatial Reasoning in Vision-Language Models
442. ICLR 2026Interpretable 3D Neural Object Volumes for Robust Conceptual Reasoning
443. ICLR 2026Is In-Context Learning Learning?
444. ICLR 2026Is it Thinking or Cheating? Detecting Implicit Reward Hacking by Measuring Reasoning Effort
445. ICLR 2026Is the Reversal Curse a Binding Problem? Uncovering Limitations of Transformers from a Basic Generalization Failure
446. ICLR 2026It's All Connected: A Journey Through Test-Time Memorization, Attentional Bias, Retention, and Online Optimization
447. ICLR 2026IV-Bench: A Benchmark for Image-Grounded Video Perception and Reasoning in Multimodal LLMs
448. ICLR 2026J1: Incentivizing Thinking in LLM-as-a-Judge via Reinforcement Learning
449. ICLR 2026Journey to the Centre of Cluster: Harnessing Interior Nodes for A/B Testing under Network Interference
450. ICLR 2026KaVa: Latent Reasoning via Compressed KV-Cache Distillation

常见问题

2023–2026 年四大会议一共有多少篇推理模型论文？

共 3,512 篇，按年份分别是 2023 年 470 篇、2024 年 311 篇、2025 年 932 篇、2026 年 1,799 篇。

NeurIPS 2026 有多少篇推理模型论文？

NeurIPS 2026 有 361 篇推理模型论文，上一届是 500 篇。

ICLR 2026 有多少篇推理模型论文？

ICLR 2026 有 652 篇推理模型论文，上一届是 203 篇。

ICML 2026 有多少篇推理模型论文？

ICML 2026 有 582 篇推理模型论文，上一届是 159 篇。

CVPR 2026 有多少篇推理模型论文？

CVPR 2026 有 204 篇推理模型论文，上一届是 70 篇。

哪个会议的推理模型论文最多？

NeurIPS，2023–2026 年共 1,253 篇。