Yangsibo Huang
23 papers · 2020–2025 · 5 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 ↓+8 more ↓ Show less ↑
🗺️ Taxonomy Completionist (34) 🏃 Academic Marathon · Sprint 🌍 Conference Polyglot · Silver 🌉 Interdisciplinary Bridge · Bronze 🐣 Hot Topic Early Bird
👑
Triple Crown
👥
Mega-Team · Squad
🤝
Dynamic Duo
(10)
🧬
Topic Evolution
🔥
Unstoppable · Bronze
💎
Century Club · Bronze
🗃️
Keyword Collector
(109)
⚡
Prolific Year · Prolific
Top co-authors
Research topics
Artificial Intelligence
(17)
Core AI
(16)
Machine Learning
(14)
Application Areas
(10)
Natural Language Processing
(7)
Resources & Methods
(4)
Deep Learning
(3)
Learning Paradigms · Artificial Intelligence
(3)
Learning Paradigms · Machine Learning
(3)
Learning Types
(3)
Optimization & Theory
(3)
Computer Vision
(2)
Generation · Natural Language Processing
(2)
Generation · Computer Vision
(2)
Models
(2)
Keywords
large language model
(5)
federated learning
(4)
differential privacy
(3)
machine unlearning
(2)
language model
(2)
data privacy
(2)
red teaming
(2)
gradient attack
(2)
privacy leakage
(2)
image generation
(2)
adversarial robustness
(1)
mathematical reasoning
(1)
text reconstruction
(1)
distributed learning
(1)
visual question answering
(1)
benchmark evaluation
(1)
logical reasoning
(1)
harmful content
(1)
video generation
(1)
stochastic gradient descent
(1)
Papers
Exploring and Mitigating Adversarial Manipulation of Voting-Based Leaderboards
ICML 2025
Fantastic Copyrighted Beasts and How (Not) to Generate Them
ICLR 2025
GMValuator: Similarity-based Data Valuation for Generative Models
ICLR 2025
MATH-Perturb: Benchmarking LLMs’ Math Reasoning Abilities against Hard Perturbations
ICML 2025
MUSE: Machine Unlearning Six-Way Evaluation for Language Models
ICLR 2025
On Evaluating the Durability of Safeguards for Open-Weight LLMs
ICLR 2025
On Memorization of Large Language Models in Logical Reasoning
IJCNLP 2025
Scaling Laws for Differentially Private Language Models
ICML 2025
SORRY-Bench: Systematically Evaluating Large Language Model Safety Refusal
ICLR 2025
Unlearn and Burn: Adversarial Machine Unlearning Requests Destroy Model Accuracy
ICLR 2025
Assessing the Brittleness of Safety Alignment via Pruning and Low-Rank Modifications
ICML 2024
Catastrophic Jailbreak of Open-source LLMs via Exploiting Generation
ICLR 2024
ConceptMix: A Compositional Image Generation Benchmark with Controllable Difficulty
NeurIPS 2024
Detecting Pretraining Data from Large Language Models
ICLR 2024
Evaluating Copyright Takedown Methods for Language Models
NeurIPS 2024
LabelDP-Pro: Learning with Label Differential Privacy via Projections
ICLR 2024
Position: A Safe Harbor for AI Evaluation and Red Teaming
ICML 2024
Privacy Implications of Retrieval-Based Language Models
EMNLP 2023
Sparsity-Preserving Differentially Private Training of Large Embedding Models
NeurIPS 2023
Recovering Private Text in Federated Learning of Language Models
NeurIPS 2022
Evaluating Gradient Inversion Attacks and Defenses in Federated Learning
NeurIPS 2021
InstaHide: Instance-hiding Schemes for Private Distributed Learning
ICML 2020
TextHide: Tackling Data Privacy in Language Understanding Tasks
EMNLP 2020