Stephen Roller
Artificial Intelligence · 2012–2025
A researcher in the Science corpus: 19 papers; trusts 1,143 researchers and is trusted by 783 researchers here, as inferred from the citation graph.
Subfields
- Artificial Intelligence — 744 citations into it; #1,175 of 36,961 by trust here, #938 by citations
- Natural Language Processing — 529 citations into it; #418 of 5,725 by trust here, #340 by citations
- Computer Vision and Pattern Recognition — 217 citations into it; #4,561 of 21,973 by trust here, #1,975 by citations
- Large Language Models — 127 citations into it; #734 of 810 by trust here, #239 by citations
- Evaluation & Benchmarks — 102 citations into it; #494 of 3,155 by trust here, #1,011 by citations
- Safety, Robustness & Interpretability — 57 citations into it
- Model Architectures — 55 citations into it
- Information Systems — 51 citations into it; #5,769 of 9,233 by trust here, #4,071 by citations
Whom Stephen Roller trusts overall
Their strongest outbound trust, strongest first, inferred from their citations.
- Jason Aaron Edward Weston
- Yoshua Bengio
- Geoffrey E. Hinton
- Diederik P. Kingma
- Ilya Sutskever
- Alex Krizhevsky
- Emily Dinan
- Oriol Vinyals
- Kaiming He
- Yann LeCun
Who trusts Stephen Roller overall
- Kurt Shuster
- Eric Michael Smith
- Emily Dinan
- Jason Aaron Edward Weston
- Jing Xu
- Y-Lan Boureau
- Da Ju
- Margaret Li
- Mojtaba Komeili
- Arthur Szlam
Papers from Stephen Roller's perspective
Ordered by how strongly the researchers Stephen Roller trusts endorse each paper. Same corpus, a different order for every researcher.
- Deep Residual Learning for Image Recognition (Computer Vision and Pattern Recognition)
- ImageNet classification with deep convolutional neural networks
- Adam: A Method for Stochastic Optimization (International Conference on Learning Representations)
- Attention is All you Need (Neural Information Processing Systems)
- Long Short-Term Memory
- BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding (North American Chapter of the Association for Computational Linguistics)
- ImageNet: A large-scale hierarchical image database
- Gradient-based learning applied to document recognition
- Sequence to Sequence Learning with Neural Networks (Neural Information Processing Systems)
- Neural Machine Translation by Jointly Learning to Align and Translate (International Conference on Learning Representations)
Their papers
- OPT: Open Pre-trained Transformer Language Models (2022) — 4,844 citations
- Wizard of Wikipedia: Knowledge-Powered Conversational agents (2018) — 1,031 citations
- Neural Text Generation with Unlikelihood Training (2019) — 727 citations
- Recipes for Building an Open-Domain Chatbot (2020) — 1,145 citations
- What makes a good conversation? How controllable attributes affect human judgments (2019) — 327 citations
- Hash Layers For Large Sparse Models (2021) — 301 citations
- ACUTE-EVAL: Improved Dialogue Evaluation with Optimized Questions and Multi-turn Comparisons (2019) — 184 citations
- Language Models that Seek for Knowledge: Modular Search & Generation for Dialogue and Prompt Completion (2022) — 150 citations
- Don’t Say That! Making Inconsistent Dialogue Unlikely with Unlikelihood Training (2019) — 212 citations
- The Dialogue Dodecathlon: Open-Domain Knowledge and Image Grounded Conversational Agents (2019) — 87 citations
- BlenderBot 3: a deployed conversational agent that continually learns to responsibly engage (2022) — 297 citations
- Open-Domain Conversational Agents: Current Progress, Open Problems, and Future Directions (2020) — 62 citations
- A Theory on Adam Instability in Large-Scale Machine Learning (2023) — 56 citations
- Human-level play in the game of Diplomacy by combining language models with strategic reasoning (2022) — 592 citations
- Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities (2025) — 3,609 citations
- Scaling Laws for Generative Mixed-Modal Language Models (2023) — 159 citations
- Human Evaluation of Conversations is an Open Problem: comparing the sensitivity of various methods for evaluating dialogue agents (2022) — 87 citations
- MGNC-CNN: A Simple Approach to Exploiting Multiple Word Embeddings for Sentence Classification (2016) — 110 citations
- Supervised Text-based Geolocation Using Language Models on an Adaptive Grid (2012) — 233 citations