Qiang Sun
Associate Professor
Studies statistical foundations of generative and trustworthy AI.
Qiang Sun's lab at Toronto works on learning theory & optimization, trustworthy & safe ML and interpretability, evaluation & safety of LMs. 23 papers since 2023, 17 at top venues, mostly ICML and ICLR. 39 alumni, most often to Amazon and Microsoft. Applications close .
At a glance
- papers, 2023–2026
- 6/ year
- at T1 venues
- 74%
- PhD students
- 0
- MS students
- 0
- alumni listed
- 39
Papers by yearfrom the lab homepage, 2023–2026
- 8 papers2026 · T1 7 · collecting
- 7 papers2025 · T1 4
- 4 papers2024 · T1 4
- 4 papers2023 · T1 2
T1 venuesOtherVenues: ICML, ICLR, NeurIPS, TMLR, UAI
Research focuspaper titles that use each topic's words
- Learning theory & optimization4 papers
- Trustworthy & safe ML0 papers
- Interpretability, evaluation & safety of LMs0 papers
- statistical0 papers
- studies statistical0 papers
Paper list
From the lab homepage.
20268 papers
- Annales de l'Institut Henri PoincaréBernstein's inequality for general Markov chains (opens in a new tab)
- CVPR (T1)CamDirector: Towards Long-Term Coherent Video Trajectory Editing (opens in a new tab)
- ICLR (T1)DeCo-DETR: Decoupled Cognition DETR for efficient Open-Vocabulary Object Detection
- ICML (T1)Feature bagging improves feature stability (opens in a new tab)
- ICML (T1)Mixtures of geodesic factor analyzers on Riemannian homogeneous spaces (opens in a new tab)
- JMLR (T1)Online generalized sparse regression: How does overparametrization help? (opens in a new tab)
- TMLR (T1)Order from Chaos: Physical World Understanding from Glitchy Gameplay Videos (opens in a new tab)
- ICML (T1)Why Self-Distillation Helps and Hurts: Denoising vs. Signal Forgetting (opens in a new tab)
20257 papers
- UAI (T1)Corruption-robust variance-aware algorithms for generalized linear bandits under heavy-tailed rewards (opens in a new tab)
- SIAM Journal on Mathematics of Data ScienceEnsemble linear interpolators: The role of ensembling (opens in a new tab)
- ICLR (T1)Intervening anchor token: Decoding strategy in alleviating hallucinations for MLLMs (opens in a new tab)
- NeurIPS (T1)PCA++: How uniformity induces robustness to background noise in contrastive learning (opens in a new tab)
- CV2 workshop @ CVPRPhysGame: Uncovering physical commonsense violations in gameplay videos (opens in a new tab)
- Electronic Journal of StatisticsResistant convex clustering: How does fusion penalty enhance robustness? (opens in a new tab)
- ICML (T1)UltraTWD: Optimizing ultrametric trees for tree-Wasserstein distance
20244 papers
- UAI (T1)Gradient descent in matrix factorization: Understanding large initialization
- TPAMI (T1)Quadratic matrix factorization with applications to manifold learning (opens in a new tab)
- ICLR (T1)Rethinking the uniformity metric in self-supervised learning (opens in a new tab)
- TMLR (T1)Variance-aware robust decision making with linear function approximation under heavy-tailed awards (opens in a new tab)
20234 papers
- Information and InferenceAn analysis of classical multidimensional scaling with applications to clustering
- NeurIPS (T1)Directional diffusion models for graph representation learning (opens in a new tab)
- Linear Algebra and its ApplicationsRestricted Riemannian geometry for positive semidefinite matrices (opens in a new tab)
- ICML (T1)Sketched ridgeless linear regression: The role of downsampling (opens in a new tab)
Where alumni went39 listed, 15% to industry
Industry 3Academia 17
Most often to
- Amazon
- Microsoft
- Shanghai AI Lab
Also to
- University of Toronto
- University of Oxford
- University of Pennsylvania
Apply to TorontoFall 2027
Computer Science MSc/PhD
- Round
- Single round (Dec 1)
- Opens
- —
- GRE
- optional
- English
- TOEFL/IELTS if applicable
- Fee
- CAD 135
Confirmed on the program page.
Similar labsnearest by research summary
- USUMD
Han Shao
- Trustworthy & safe ML
- CNTsinghua
Zhixuan Fang
- Reasoning & neuro-symbolic AI
- SGNTU
Han Yu
- Trustworthy & safe ML
- SGNUS
Kenji Kawaguchi
- Learning theory & optimization
- CNSJTU
Nanyang Ye
- Learning theory & optimization
- SGNTU
Atsushi Nitanda
- 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.
Torontocloses