Geoffrey Irving

Computer Science · Physical Sciences

30h-index29.5kcitations56works0.02yr avg
OpenAlex↗Google Scholar↗

Accepting Students?

No reports yet. Know if this professor is taking students?

Research Topics

Topic Modeling(20), Natural Language Processing Techniques(19), Computer Graphics and Visualization Techniques(11), 3D Shape Modeling and Analysis(7), Parallel Computing and Optimization Techniques(5)

Publications55 total

Avoiding Obfuscation with Prover-Estimator Debate
arXiv (Cornell University)·2025Open Access
Scalable AI Safety via Doubly-Efficient Debate
arXiv (Cornell University)·2023· 1 citedOpen Access
Does Circuit Analysis Interpretability Scale? Evidence from Multiple Choice Capabilities in Chinchilla
arXiv (Cornell University)·2023· 6 citedOpen Access
Accelerating Large Language Model Decoding with Speculative Sampling
arXiv (Cornell University)·2023· 26 citedOpen Access
Solving math word problems with process- and outcome-based feedback
arXiv (Cornell University)·2022· 21 citedOpen Access
Fine-Tuning Language Models via Epistemic Neural Networks
arXiv (Cornell University)·2022· 4 citedOpen Access
Improving alignment of dialogue agents via targeted human judgements
arXiv (Cornell University)·2022· 130 citedOpen Access
Taxonomy of Risks posed by Language Models
2022 ACM Conference on Fairness, Accountability, and Transparency·2022· 502 citedOpen Access
Characteristics of Harmful Text: Towards Rigorous Benchmarking of Language Models
arXiv (Cornell University)·2022· 12 citedOpen Access
Teaching language models to support answers with verified quotes
arXiv (Cornell University)·2022· 52 citedOpen Access
Uncertainty Estimation for Language Reward Models
arXiv (Cornell University)·2022· 7 citedOpen Access
Red Teaming Language Models with Language Models
arXiv (Cornell University)·2022· 17 citedOpen Access
Red Teaming Language Models with Language Models
2022· 223 citedOpen Access
Scaling Language Models: Methods, Analysis & Insights from Training Gopher
arXiv (Cornell University)·2021· 242 citedOpen Access
Ethical and social risks of harm from Language Models
arXiv (Cornell University)·2021· 71 citedOpen Access
Improving language models by retrieving from trillions of tokens
arXiv (Cornell University)·2021· 295 citedOpen Access
Alignment of Language Agents
arXiv (Cornell University)·2021· 41 citedOpen Access
SymbioticLab/tensorflow-salus: Release code for MLSys20
Zenodo (CERN European Organization for Nuclear Research)·2020Open Access
Fine-Tuning Language Models from Human Preferences
arXiv (Cornell University)·2019· 382 citedOpen Access
AI Safety Needs Social Scientists
Distill·2019· 46 citedOpen Access
page 1 of 3Next →

Frequent Co-authors

Ronald Fedkiw(10), Oriol Vinyals(7), Martı́n Abadi(6), Paul Barham(6), Andy Davis(6), Matthieu Devin(6), Sanjay Ghemawat(6), Michael Isard(6), Manjunath Kudlur(6), Josh Levenberg(6), Rajat Monga(6), Sherry Moore(6), Derek G. Murray(6), Benoit Steiner(6), Paul A. Tucker(6), Pete Warden(6), Laura Weidinger(6), Jonathan Uesato(6), Amelia Glaese(6), Iason Gabriel(6)