PaperHub

Jason D. Lee

~Jason_D._Lee1

41
论文总数
20.5
年均投稿
6.3
平均评分
接收情况32/41
会议分布
ICLR
24
NeurIPS
13
ICML
4

发表论文 (41 篇)

202525

7.3
3

Understanding Factual Recall in Transformers via Associative Memories

ICLR 2025Spotlight
7.3
4

Learning Orthogonal Multi-Index Models: A Fine-Grained Information Exponent Analysis

NeurIPS 2025Poster
6.0
3

Learning Orthogonal Multi-Index Models: A Fine-Grained Information Exponent Analysis

ICLR 2025Rejected
8.7
4

The Generative Leap: Tight Sample Complexity for Efficiently Learning Gaussian Multi-Index Models

NeurIPS 2025Spotlight
7.3
3

Transformers Learn to Implement Multi-step Gradient Descent with Chain of Thought

ICLR 2025Spotlight
6.1
4

Minimax Optimal Regret Bound for Reinforcement Learning with Trajectory Feedback

ICML 2025Poster
6.8
4

Deployment Efficient Reward-Free Exploration with Linear Function Approximation

NeurIPS 2025Poster
6.5
4

Transformers Provably Learn Two-Mixture of Linear Classification via Gradient Flow

ICLR 2025Poster
5.5
6

Minimax Optimal Regret Bound for Reinforcement Learning with Trajectory Feedback

ICLR 2025Rejected
7.0
3

Metastable Dynamics of Chain-of-Thought Reasoning: Provable Benefits of Search, RL and Distillation

ICML 2025Poster
6.8
5

Learning Hierarchical Polynomials of Multiple Nonlinear Features

ICLR 2025Poster
7.3
4

Accelerating RL for LLM Reasoning with Optimal Advantage Regression

NeurIPS 2025Poster
7.0
4

Exploiting Structure in Offline Multi-Agent RL: The Benefits of Low Interaction Rank

ICLR 2025Poster
6.6
4

Rethinking Addressing in Language Models via Contextualized Equivariant Positional Encoding

ICML 2025Poster
6.4
5

Correcting the Mythos of KL-Regularization: Direct Alignment without Overoptimization via Chi-Squared Preference Optimization

ICLR 2025Spotlight
4.4
4

Discrepancies are Virtue: Weak-to-Strong Generalization through Lens of Intrinsic Dimension

ICML 2025Poster
7.8
4

Emergence and scaling laws in SGD learning of shallow neural networks

NeurIPS 2025Poster
6.0
4

Rethinking Addressing in Language Models via Contextualized Equivariant Positional Encoding

ICLR 2025Rejected
4.8
4

Deployment Efficient Reward-Free Exploration with Linear Function Approximation

ICLR 2025Rejected
7.5
5

What Makes a Reward Model a Good Teacher? An Optimization Perspective

NeurIPS 2025Spotlight
7.0
5

Understanding Optimization in Deep Learning with Central Flows

ICLR 2025Poster
6.0
4

Watermarking using Semantic-aware Speculative Sampling: from Theory to Practice

ICLR 2025Rejected
6.8
4

What One Cannot, Two Can: Two-Layer Transformers Provably Represent Induction Heads on Any-Order Markov Chains

NeurIPS 2025Spotlight
4.5
4

Task Diversity Shortens the ICL Plateau

ICLR 2025Rejected
6.5
4

Regressing the Relative Future: Efficient Policy Optimization for Multi-turn RLHF

ICLR 2025Poster

202416