Marvin Li
4 papers · 2023–2026 · 3 venues · across top CS/AI venues
Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.
Achievements
Jump to papers ↓
🌍
Conference Polyglot · Bronze
🌉
Interdisciplinary Bridge · Silver
Top co-authors
Research topics
Keywords
critical windows
(2)
diffusion model
(2)
large language model
(2)
autoregressive model
(1)
membership inference attack
(1)
privacy evaluation
(1)
benchmark construction
(1)
parameter perturbation
(1)
model memorization
(1)
feature emergence
(1)
training data leakage
(1)
log-concave densities
(1)
localization
(1)
distribution
(1)
hierarchical samplers
(1)
privacy attack
(1)
inter-group separation
(1)
hessian matrix
(1)
generative model
(1)
Papers
CheckMIABench: Firm Foundations For Membership Inference Attacks on Language Models
ACL 2026
Blink of an eye: a simple theory for feature localization in generative models
ICML 2025
Critical windows: non-asymptotic theory for feature emergence in diffusion models
ICML 2024
MoPe: Model Perturbation based Privacy Attacks on Language Models
EMNLP 2023