Conference on Neural Information Processing Systems
Fraction of papers covering each topic over time
Topics over-represented at NeurIPS vs. the field average (2025)
| Year | Title | Citations | Links |
|---|---|---|---|
| 2020 | Language Models are Few-Shot Learners | 54,053 | S2 · arXiv |
| 2019 | PyTorch: An Imperative Style, High-Performance Deep Learning Library | 49,641 | S2 · arXiv |
| 2020 | Denoising Diffusion Probabilistic Models | 27,025 | S2 · arXiv |
| 2022 | Training language models to follow instructions with human feedback | 18,387 | S2 · arXiv |
| 2022 | Chain of Thought Prompting Elicits Reasoning in Large Language Models | 15,489 | S2 · arXiv |
| 2020 | Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks | 11,260 | S2 · arXiv |
| 2021 | Diffusion Models Beat GANs on Image Synthesis | 10,681 | S2 · arXiv |
| 2019 | XLNet: Generalized Autoregressive Pretraining for Language Understanding | 9,165 | S2 · arXiv |
| 2020 | Bootstrap Your Own Latent: A New Approach to Self-Supervised Learning | 8,057 | S2 · arXiv |
| 2023 | Visual Instruction Tuning | 7,869 | S2 · arXiv |
| 2022 | Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding | 7,698 | S2 · arXiv |
| 2020 | wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations | 7,606 | S2 · arXiv |
| 2021 | SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers | 7,313 | S2 · arXiv |
| 2023 | Direct Preference Optimization: Your Language Model is Secretly a Reward Model | 7,231 | S2 · arXiv |
| 2023 | Judging LLM-as-a-judge with MT-Bench and Chatbot Arena | 7,047 | S2 · arXiv |
| 2022 | Large Language Models are Zero-Shot Reasoners | 6,370 | S2 · arXiv |
| 2020 | Supervised Contrastive Learning | 5,701 | S2 · arXiv |
| 2022 | Flamingo: a Visual Language Model for Few-Shot Learning | 5,057 | S2 · arXiv |
| 2019 | Generative Modeling by Estimating Gradients of the Data Distribution | 5,025 | S2 · arXiv |
| 2020 | Unsupervised Learning of Visual Features by Contrasting Cluster Assignments | 4,736 | S2 · arXiv |