Yarin Gal
Associate Professor of Machine Learning
Leads the OATML group on Bayesian deep learning, uncertainty estimation and safe, robust AI.
Yarin Gal's lab at Oxford works on probabilistic & causal inference and trustworthy & safe ML. 38 papers since 2023, 22 at top venues, mostly ICML and NeurIPS. 12 PhD students. 20 alumni. Applications close .
At a glance
- papers, 2023–2026
- 10/ year
- at T1 venues
- 58%
- PhD students
- 12
- MS students
- 0
- alumni listed
- 20
Papers by yearfrom the lab homepage, 2023–2026
- 9 papers2026 · T1 8 · collecting
- 7 papers2025 · T1 3
- 7 papers2024 · T1 2
- 15 papers2023 · T1 9
T1 venuesOtherVenues: ICML, NeurIPS, ICLR, AAAI, ACL
Research focuspaper titles that use each topic's words
- Probabilistic & causal inference8 papers
- Trustworthy & safe ML0 papers
- bayesian deep0 papers
- uncertainty estimation0 papers
- safe0 papers
Paper list
From the lab homepage.
20269 papers
- ICML (T1)Building Reliable Long-Form Generation via Hallucination Rejection Sampling (opens in a new tab)
- ICML (T1)MADE: Benchmark Environments for Closed-Loop Materials Discovery (opens in a new tab)
- ACL (T1)MEDSYN: Benchmarking Multi-EviDence SYNthesis in Complex Clinical Cases for Multimodal Large Language Models (opens in a new tab)
- ICML (T1)OTora: A Unified Red Teaming Framework for Reasoning-Level Denial-of-Service in LLM Agents (opens in a new tab)
- ECML-PKDDPRISM-CTG: A Foundation Model for Cardiotocography Analysis with Multi-View SSL (opens in a new tab)
- ICLR (T1)Rethinking Data Curation in LLM Training: Online Reweighting Offers Better Generalization than Offline Methods (opens in a new tab)
- ICML (T1)Richer Bayesian Last Layers with Subsampled NTK Features (opens in a new tab)
- ICLR (T1)Stabilizing Policy Gradients for Sample-Efficient Reinforcement Learning in LLM Reasoning (opens in a new tab)
- AAAI (T1)STACK: Adversarial Attacks on LLM Safeguard Pipelines (opens in a new tab)
20257 papers
- PNAS (T1)Bridging the Human–AI Knowledge Gap through Concept Discovery and Transfer in AlphaZero (opens in a new tab)
- NeurIPS WorkshopEnergy Landscapes Enable Reliable Abstention in Retrieval-Augmented Large Language Models for Healthcare (opens in a new tab)
- The Lancet Obstetrics, Gynaecology, & Women’s HealthReducing Large Language Model Safety Risks in Women's Health using Semantic Entropy (opens in a new tab)
- ICML (T1)Rethinking Aleatoric and Epistemic Uncertainty (opens in a new tab)
- Published at ICMLSliding Puzzles Gym: A Scalable Benchmark for State Representation in Visual Reinforcement Learning (opens in a new tab)
- NeurIPS (T1)Temporal-Difference Variational Continual Learning (opens in a new tab)
- Nature and Gaming Intelligence@MSRWorld and Human Action Models towards gameplay ideation (opens in a new tab)
20247 papers
- NeurIPS (T1)Deep Bayesian Active Learning for Preference Modeling in Large Language Models (opens in a new tab)
- ICLR (T1)In-Context Learning Learns Label Relationships but Is Not Conventional Learning (opens in a new tab)
- Nature BiotechnologyMachine learning for functional protein design (opens in a new tab)
- International Conference on Artificial Intelligence and Statistics (AISTATS)Making Better Use of Unlabelled Data in Bayesian Active Learning (opens in a new tab)
- ICLR WorkshopPre-trained Text-to-Image Diffusion Models Are Versatile Representation Learners for Control (opens in a new tab)
- ICML WorkshopSemantic Entropy Probes: Robust and Cheap Hallucination Detection in LLMs (opens in a new tab)
- ICML WorkshopVariational Inference Failures Under Model Symmetries: Permutation Invariant Posteriors for Bayesian Neural Networks (opens in a new tab)
202315 papers
- TMLR (T1)Active Acquisition for Multimodal Temporal Data: A Challenging Decision-Making Task (opens in a new tab)
- ICML (T1)B-Learner: Quasi-Oracle Bounds on Heterogeneous Causal Effects Under Hidden Confounding (opens in a new tab)
- Structured Probabilistic Inference & Generative Modeling workshop, ICMLBatchGFN: Generative Flow Networks for Batch Active Learning (opens in a new tab)
- CLeaRCan Active Sampling Reduce Causal Confusion in Offline Reinforcement Learning? (opens in a new tab)
- ICML (T1)Differentiable Multi-Target Causal Bayesian Experimental Design (opens in a new tab)
- ICML (T1)DiscoBAX - Discovery of optimal intervention sets in genomic experiment design (opens in a new tab)
- ICML (T1)Drug Discovery under Covariate Shift with Domain-Informed Prior Distributions over Functions (opens in a new tab)
- Machine Learning and the Physical Sciences workshop, NeurIPSHigh-Cadence Thermospheric Density Estimation enabled by Machine Learning on Solar Imagery (opens in a new tab)
- Science (T1)Managing extreme AI risks amid rapid progress (opens in a new tab)
- Journal of Photogrammetry and Remote SensingMulti-Spectral Multi-Image Super-Resolution of Sentinel-2 with Radiometric Consistency Losses and Its Effect on Building Delineation (opens in a new tab)
- International Conference on Artificial Intelligence and Statistics (AISTATS)Prediction-Oriented Bayesian Active Learning (opens in a new tab)
- NeurIPS (T1)ProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design (opens in a new tab)
Where alumni went20 listed
Apply to Oxford2027-28
DPhil in Computer Science
- Round
- 2027-28 entry (January round)
- Opens
- —
- GRE
- See program page
- English
- See program page
- Fee
- See program page
Not yet confirmed for this intake.
Similar labsnearest by research summary
- SGNTU
Lihui Chen
- Recognition, detection & segmentation
- USMichigan
Michael Wellman
- Reasoning & neuro-symbolic AI
- GBEdinburgh
Timothy Hospedales
- Efficient training & inference
- USUIUC
Maxim Raginsky
- Learning theory & optimization
- USMichigan
Michał Dereziński
- Learning theory & optimization
- USCMU
Sarah Cen
- Learning theory & optimization
Your notes
Save this lab to keep notes.
Data
Checked against the lab homepage.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Lab homepage
- Alumni
- Lab homepage
Tiers are venue classifications, not lab rankings.
Oxfordcloses