Hari Sundaram
8 papers · 2022–2025 · 6 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓+1 more ↓ Show less ↑
π Interdisciplinary Bridge π§ Keyword Pioneer π Conference Polyglot (6) π Cross-Pollinator (10) π Renaissance Researcher (5)
πΊοΈ
Taxonomy Completionist
(20)
Conferences
EACL (2)
EMNLP (2)
ACL (1)
AISTATS (1)
ICLR (1)
ICML (1)
Top co-authors
Keywords
language model
(2)
large language model
(2)
text generation
(2)
style transfer
(1)
constrained generation
(1)
batch normalization
(1)
toxicity detection
(1)
code generation
(1)
text style transfer
(1)
ai safety
(1)
automatic differentiation
(1)
spectral norm
(1)
pairwise comparison
(1)
sparse autoencoder
(1)
language generation
(1)
causal intervention
(1)
residual stream
(1)
jailbreak defense
(1)
spectral norm computation
(1)
convolutional layer
(1)
Papers
Not All Wrong is Bad: Using Adversarial Examples for Unlearning
ICML 2025
Breaking Bad Tokens: Detoxification of LLMs Using Sparse Autoencoders
EMNLP 2025
Training Robust Ensembles Requires Rethinking Lipschitz Continuity
ICLR 2025
Advancing Precise Outline-Conditioned Text Generation with Task Duality and Explicit Outline Control
EACL 2024
CodeScope: An Execution-based Multilingual Multitask Multidimensional Benchmark for Evaluating LLMs on Code Understanding and Generation
ACL 2024
Spectrum Extraction and Clipping for Implicitly Linear Layers
AISTATS 2024
CEV-LM: Controlled Edit Vector Language Model for Shaping Natural Language Generations
EACL 2024
Audience-Centric Natural Language Generation via Style Infusion
EMNLP 2022