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 1,253 篇，第 15 / 26 页

701. NeurIPS 2025Reinforcing Spatial Reasoning in Vision-Language Models with Interwoven Thinking and Visual Drawing
702. NeurIPS 2025Reinforcing the Diffusion Chain of Lateral Thought with Diffusion Language Models
703. NeurIPS 2025Representation Consistency for Accurate and Coherent LLM Answer Aggregation
704. NeurIPS 2025Representation-Level Counterfactual Calibration for Debiased Zero-Shot Recognition
705. NeurIPS 2025ReSearch: Learning to Reason with Search for LLMs via Reinforcement Learning
706. NeurIPS 2025ReservoirTTA: Prolonged Test-time Adaptation for Evolving and Recurring Domains
707. NeurIPS 2025Restricted Spectral Gap Decomposition for Simulated Tempering Targeting Mixture Distributions
708. NeurIPS 2025Rethinking Fine-Tuning when Scaling Test-Time Compute: Limiting Confidence Improves Mathematical Reasoning
709. NeurIPS 2025Rethinking Optimal Verification Granularity for Compute-Efficient Test-Time Scaling
710. NeurIPS 2025Revisiting Reinforcement Learning for LLM Reasoning from A Cross-Domain Perspective
711. NeurIPS 2025Reviving DSP for Advanced Theorem Proving in the Era of Reasoning Models
712. NeurIPS 2025Reward Reasoning Models
713. NeurIPS 2025RHYTHM: Reasoning with Hierarchical Temporal Tokenization for Human Mobility
714. NeurIPS 2025Right for the Right Reasons: Avoiding Reasoning Shortcuts via Prototypical Neurosymbolic AI
715. NeurIPS 2025Right Question is Already Half the Answer: Fully Unsupervised LLM Reasoning Incentivization
716. NeurIPS 2025RL Tango: Reinforcing Generator and Verifier Together for Language Reasoning
717. NeurIPS 2025RoboRefer: Towards Spatial Referring with Reasoning in Vision-Language Models for Robotics
718. NeurIPS 2025Robot-R1: Reinforcement Learning for Enhanced Embodied Reasoning in Robotics
719. NeurIPS 2025Rollout Roulette: A Probabilistic Inference Approach to Inference-Time Scaling of LLMs using Particle-Based Monte Carlo Methods
720. NeurIPS 2025ROVER: Recursive Reasoning Over Videos with Vision-Language Models for Embodied Tasks
721. NeurIPS 2025RrED: Black-box Unsupervised Domain Adaptation via Rectifying-reasoning Errors of Diffusion
722. NeurIPS 2025rStar-Coder: Scaling Competitive Code Reasoning with a Large-Scale Verified Dataset
723. NeurIPS 2025RTV-Bench: Benchmarking MLLM Continuous Perception, Understanding and Reasoning through Real-Time Video
724. NeurIPS 2025S-GRPO: Early Exit via Reinforcement Learning in Reasoning Models
725. NeurIPS 2025SAFEPATH: Preventing Harmful Reasoning in Chain-of-Thought via Early Alignment
726. NeurIPS 2025SAM-R1: Leveraging SAM for Reward Feedback in Multimodal Segmentation via Reinforcement Learning
727. NeurIPS 2025Sampling-Efficient Test-Time Scaling: Self-Estimating the Best-of-N Sampling in Early Decoding
728. NeurIPS 2025SATURN: SAT-based Reinforcement Learning to Unleash LLMs Reasoning
729. NeurIPS 2025Scalable Best-of-N Selection for Large Language Models via Self-Certainty
730. NeurIPS 2025Scaling Code-Assisted Chain-of-Thoughts and Instructions for Model Reasoning
731. NeurIPS 2025Scaling Physical Reasoning with the PHYSICS Dataset
732. NeurIPS 2025Scaling RL to Long Videos
733. NeurIPS 2025Scaling Speculative Decoding with Lookahead Reasoning
734. NeurIPS 2025Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach
735. NeurIPS 2025SCAN: Self-Denoising Monte Carlo Annotation for Robust Process Reward Learning
736. NeurIPS 2025Scent of Knowledge: Optimizing Search-Enhanced Reasoning with Information Foraging
737. NeurIPS 2025SCOUT: Teaching Pre-trained Language Models to Enhance Reasoning via Flow Chain-of-Thought
738. NeurIPS 2025scPilot: Large Language Model Reasoning Toward Automated Single-Cell Analysis and Discovery
739. NeurIPS 2025SD-VLM: Spatial Measuring and Understanding with Depth-Encoded Vision-Language Models
740. NeurIPS 2025SE-Agent: Self-Evolution Trajectory Optimization in Multi-Step Reasoning with LLM-Based Agents
741. NeurIPS 2025Search and Refine During Think: Facilitating Knowledge Refinement for Improved Retrieval-Augmented Reasoning
742. NeurIPS 2025Searching Latent Program Spaces
743. NeurIPS 2025SeePhys: Does Seeing Help Thinking? – Benchmarking Vision-Based Physics Reasoning
744. NeurIPS 2025Segment Policy Optimization: Effective Segment-Level Credit Assignment in RL for Large Language Models
745. NeurIPS 2025Self-Verifying Reflection Helps Transformers with CoT Reasoning
746. NeurIPS 2025SemCoT: Accelerating Chain-of-Thought Reasoning through Semantically-Aligned Implicit Tokens
747. NeurIPS 2025Semi-off-Policy Reinforcement Learning for Vision-Language Slow-Thinking Reasoning
748. NeurIPS 2025Sherlock: Self-Correcting Reasoning in Vision-Language Models
749. NeurIPS 2025ShorterBetter: Guiding Reasoning Models to Find Optimal Inference Length for Efficient Reasoning
750. NeurIPS 2025SIGMA: Refining Large Language Model Reasoning via Sibling-Guided Monte Carlo Augmentation

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。