供应链可信论文趋势 2023–2026

主要结论

• 供应链可信论文从 2023 年的 9 篇增长到 2026 年的 41 篇，三年间扩大到约 4.6 倍。
• NeurIPS 是发供应链可信论文最多的会议，2023–2026 年共 35 篇，占四个会议总量的 38%。
• 交叉方向里涨得最快的是驾驶仿真：同时属于两个主题的论文从 2023 年的 1 篇增至 2026 年的 5 篇。
• 在「安全与可信」的 5 个主题里，它的 2026 年论文数排第 5；同组增长最快的水印是 9.3 倍，供应链可信是 4.6 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 2026Can We Trust Item Response Theory for AI Evaluation?
• ICML 2026Who can we trust? LLM-as-a-jury for Comparative Assessment
• ICLR 2026Hey, That's My Model! Introducing Chain & Hash, An LLM Fingerprinting Technique
• CVPR 2026Trust-calibrated Collaborative Learning for Long-Tailed Visual Recognition
• NeurIPS 2025Chain-of-Model Learning for Language Model
• ICML 2025Navigating Conflicting Views: Harnessing Trust for Learning
• ICLR 2025Parameter Expanded Stochastic Gradient Markov Chain Monte Carlo
• CVPR 2025Interleaved-Modal Chain-of-Thought
• NeurIPS 2024Chain-of-Thought Unfaithfulness as Disguised Accuracy
• ICML 2024Implicit meta-learning may lead language models to trust more reliable sources
• ICLR 2024Chain of Log-Concave Markov Chains
• CVPR 2024CoG-DQA: Chain-of-Guiding Learning with Large Language Models for Diagram Question Answering
• NeurIPS 2023Supply-Side Equilibria in Recommender Systems
• ICML 2023Synthetic Prompting: Generating Chain-of-Thought Demonstrations for Large Language Models
• ICLR 2023Rethinking skip connection model as a learnable Markov chain
• NeurIPS 2026From Infrastructure to Interface, the AI Value Chain Drives LLM Homogenization
• ICML 2026Covariance estimation using Markov chain Monte Carlo
• ICLR 2026Output Supervision Can Obfuscate the Chain of Thought
• CVPR 2026Cut to the Chase Training-free Multimodal Summarization via Chain-of-Events
• NeurIPS 2025You Can Trust Your Clustering Model: A Parameter-free Self-Boosting Plug-in for Deep Clustering

全部供应链可信论文

共 91 篇，第 1 / 2 页

1. CVPR 2026CaST-Bench Benchmarking Causal Chain-Grounded Spatio-Temporal Reasoning for Video Question Answering
2. CVPR 2026Chain of World World Model Thinking in Latent Motion
3. CVPR 2026Cut to the Chase Training-free Multimodal Summarization via Chain-of-Events
4. CVPR 2026Fuel Gauge Estimating Chain-of-Thought Length Ahead of Time in Large
5. CVPR 2026Grounded Chain-of-Thought for Multimodal Large Language Models
6. CVPR 2026Trust-calibrated Collaborative Learning for Long-Tailed Visual Recognition
7. CVPR 2026Unleashing the Power of Chain-of-Prediction for Monocular 3D Object Detection
8. ICLR 2026Emergence of Superposition: Unveiling the Training Dynamics of Chain of Continuous Thought
9. ICLR 2026Hey, That's My Model! Introducing Chain & Hash, An LLM Fingerprinting Technique
10. ICLR 2026Hidden Patterns in Chain-of-Thought Reasoning
11. ICLR 2026In Agents We Trust, but Who Do Agents Trust? Latent Preferences Steer LLM Generations
12. ICLR 2026Long Chain-of-Thought Reasoning Across Languages
13. ICLR 2026Output Supervision Can Obfuscate the Chain of Thought
14. ICLR 2026Pushing on Multilingual Reasoning Models with Language-Mixed Chain-of-Thought
15. ICLR 2026Scalable Chain of Thoughts via Elastic Reasoning
16. ICML 2026Chain-of-Thought Gradient Descent
17. ICML 2026Covariance estimation using Markov chain Monte Carlo
18. ICML 2026Reasoning about Reasoning: BAPO Bounds on Chain-of-Thought Token Complexity in LLMs
19. ICML 2026The Expressive Power of Low Precision Softmax Transformers with (Summarized) Chain-of-Thought
20. ICML 2026Thinking in Flow: A Dissipative Stabilization Operator for Robust Autoregressive Reasoning
21. ICML 2026When Can We Trust Survival Model Evaluation ?
22. ICML 2026Who can we trust? LLM-as-a-jury for Comparative Assessment
23. NeurIPS 2026A Solvable Model of Chain-of-Thought in In-Context Learning
24. NeurIPS 2026A Tight Hierarchy For Chain of Thought
25. NeurIPS 2026Beyond Suspicious Steps: Ontological Trust in Long-Horizon Agents
26. NeurIPS 2026Can We Trust Item Response Theory for AI Evaluation?
27. NeurIPS 2026Chain of Dual Structures in Transformer Attention
28. NeurIPS 2026Chain-of-Thought Is Not Explainability
29. NeurIPS 2026ComPose: When to Trust Hands for Object Pose Tracking
30. NeurIPS 2026Efficiently Representing Algorithms With Chain-of-Thought Transformers
31. NeurIPS 2026From Infrastructure to Interface, the AI Value Chain Drives LLM Homogenization
32. NeurIPS 2026Not All Routing Drift Is Harmful: Trust-Region Projection for Class-Incremental Learning
33. NeurIPS 2026Reliable Chain-of-Thought via Prefix Consistency
34. NeurIPS 2026SWE-Chain: Benchmarking Coding Agents on Chained Release-Level Package Upgrades
35. NeurIPS 2026TERMINATOR: Learning Optimal Exit Points for Early Stopping in Chain-of-Thought Reasoning
36. NeurIPS 2026Trust Guided Decision Transformer
37. NeurIPS 2026Trust Region Continual Learning as an Implicit Meta-Learner
38. NeurIPS 2026Trust Region Policy Distillation
39. NeurIPS 2026Trust Region Q Adjoint Matching
40. NeurIPS 2026When Alignment Fails: Stabilizing Cross-Dynamics RL with Prototype Trust Regions
41. NeurIPS 2026When the Same Coefficients Reach Different Places: Asymmetric Realizability in Transplanting Tokenizers across Large Language Models
42. CVPR 2025Interleaved-Modal Chain-of-Thought
43. ICLR 2025Chain-of-Thought Provably Enables Learning the (Otherwise) Unlearnable
44. ICLR 2025CoTFormer: A Chain of Thought Driven Architecture with Budget-Adaptive Computation Cost at Inference
45. ICLR 2025Don't Take Things Out of Context: Attention Intervention for Enhancing Chain-of-Thought Reasoning in Large Language Models
46. ICLR 2025More RLHF, More Trust? On The Impact of Preference Alignment On Trustworthiness
47. ICLR 2025Parameter Expanded Stochastic Gradient Markov Chain Monte Carlo
48. ICLR 2025Transformers Provably Solve Parity Efficiently with Chain of Thought
49. ICML 2025Learning to Trust Bellman Updates: Selective State-Adaptive Regularization for Offline RL
50. ICML 2025Lower Bounds for Chain-of-Thought Reasoning in Hard-Attention Transformers

常见问题

2023–2026 年四大会议一共有多少篇供应链可信论文？

共 91 篇，按年份分别是 2023 年 9 篇、2024 年 21 篇、2025 年 20 篇、2026 年 41 篇。

NeurIPS 2026 有多少篇供应链可信论文？

NeurIPS 2026 有 19 篇供应链可信论文，上一届是 8 篇。

ICLR 2026 有多少篇供应链可信论文？

ICLR 2026 有 8 篇供应链可信论文，上一届是 6 篇。

ICML 2026 有多少篇供应链可信论文？

ICML 2026 有 7 篇供应链可信论文，上一届是 5 篇。

CVPR 2026 有多少篇供应链可信论文？

CVPR 2026 有 7 篇供应链可信论文，上一届是 1 篇。

哪个会议的供应链可信论文最多？

NeurIPS，2023–2026 年共 35 篇。