Nikunj Saunshi
19 papers · 2018–2025 · 4 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓+10 more ↓ Show less ↑
π§ Keyword Pioneer π£ Hot Topic Early Bird πΊοΈ Taxonomy Completionist (12) π Interdisciplinary Bridge π Conference Polyglot (4)
π
Conference Polyglot
(4)
πΊοΈ
Taxonomy Completionist
(12)
π€
Dynamic Duo
(11)
π
Triple Crown
π
Century Club
(19)
π
Trend Setter
π
Conference Pioneer
π₯
Unstoppable
(8)
β
The Questioner
ποΈ
Keyword Collector
(53)
Conferences
ICML (8)
ICLR (6)
NIPS (3)
ACL (2)
Top co-authors
Keywords
representation learning
(6)
sample complexity
(4)
neural network optimization
(2)
self-supervised learning
(2)
inductive bia
(2)
contrastive learning
(2)
few-shot learning
(2)
model pretraining
(1)
mathematical reasoning
(1)
non-convex optimization
(1)
symbolic reasoning
(1)
domain adaptation
(1)
zero-shot learning
(1)
transfer learning
(1)
feature attribution
(1)
imitation learning
(1)
model interpretability
(1)
behavior cloning
(1)
markov decision process
(1)
continual learning
(1)
Papers
Learning to Keep a Promise: Scaling Language Model Decoding Parallelism with Learned Asynchronous Decoding
ICML 2025
Reasoning with Latent Thoughts: On the Power of Looped Transformers
ICLR 2025
Efficient stagewise pretraining via progressive subnetworks
ICLR 2025
On the Inductive Bias of Stacking Towards Improving Reasoning
NIPS 2024
Can Looped Transformers Learn to Implement Multi-step Gradient Descent for In-context Learning?
ICML 2024
Task-Specific Skill Localization in Fine-tuned Language Models
ICML 2023
Reasoning in Large Language Models Through Symbolic Math Word Problems
ACL 2023
Understanding Influence Functions and Datamodels via Harmonic Analysis
ICLR 2023
Understanding Contrastive Learning Requires Incorporating Inductive Biases
ICML 2022
New Definitions and Evaluations for Saliency Methods: Staying Intrinsic, Complete and Sound
NIPS 2022
On Predicting Generalization using GANs
ICLR 2022
A Representation Learning Perspective on the Importance of Train-Validation Splitting in Meta-Learning
ICML 2021
A Mathematical Exploration of Why Language Models Help Solve Downstream Tasks
ICLR 2021
Predicting What You Already Know Helps: Provable Self-Supervised Learning
NIPS 2021
Provable Representation Learning for Imitation Learning via Bi-level Optimization
ICML 2020
A Sample Complexity Separation between Non-Convex and Convex Meta-Learning
ICML 2020
A Theoretical Analysis of Contrastive Unsupervised Representation Learning
ICML 2019
A La Carte Embedding: Cheap but Effective Induction of Semantic Feature Vectors
ACL 2018
A Compressed Sensing View of Unsupervised Text Embeddings, Bag-of-n-Grams, and LSTMs
ICLR 2018