PaperHub

Vahab Mirrokni

~Vahab_Mirrokni2

31
论文总数
15.5
年均投稿
6.0
平均评分
接收情况23/31
会议分布
ICLR
12
NeurIPS
10
ICML
7
COLM
2

发表论文 (31 篇)

202521

6.1
4

Improving the Variance of Differentially Private Randomized Experiments through Clustering

ICML 2025Poster
6.3
4

ClusterGen: Token Generation in Sublinear Time and Memory with Clustering KV Cache

ICLR 2025Rejected
7.8
4

Titans: Learning to Memorize at Test Time

NeurIPS 2025Poster
5.3
4

Maximum Coverage in Turnstile Streams with Applications to Fingerprinting Measures

ICLR 2025Rejected
7.2
4

Procurement Auctions via Approximately Optimal Submodular Optimization

ICML 2025Spotlight
6.6
4

Maximum Coverage in Turnstile Streams with Applications to Fingerprinting Measures

ICML 2025Poster
6.4
5

Efficient Data Selection at Scale via Influence Distillation

NeurIPS 2025Poster
6.4
4

Nested Learning: The Illusion of Deep Learning Architectures

NeurIPS 2025Poster
7.8
4

Self-Boost via Optimal Retraining: An Analysis via Approximate Message Passing

NeurIPS 2025Poster
4.4
4

Synthetic Text Generation for Training Large Language Models via Gradient Matching

ICML 2025Poster
7.3
3

TRELLIS: Learning to Compress Key-Value Memory in Attention Models

COLM 2025Poster
6.8
4

MS-SSM: A Multi-Scale State Space Model for Efficient Sequence Modeling

COLM 2025Poster
3.0
4

NanoMoE: Scaling Mixture of Experts to Individual Layers for Parameter-Efficient Deep Learning

ICLR 2025Rejected
7.8
4

PiKE: Adaptive Data Mixing for Large-Scale Multi-Task Learning Under Low Gradient Conflicts

NeurIPS 2025Spotlight
7.0
4

DiSK: Differentially Private Optimizer with Simplified Kalman Filter for Noise Reduction

ICLR 2025Poster
-

Retraining with Predicted Hard Labels Provably Increases Model Accuracy

ICLR 2025desk_rejected
7.0
3

Retraining with Predicted Hard Labels Provably Increases Model Accuracy

ICML 2025Poster
5.5
3

DeepCrossAttention: Supercharging Transformer Residual Connections

ICML 2025Poster
6.0
4

Addax: Utilizing Zeroth-Order Gradients to Improve Memory Efficiency and Performance of SGD for Fine-Tuning Language Models

ICLR 2025Poster
6.3
3

Best of Both Worlds: Advantages of Hybrid Graph Sequence Models

ICML 2025Poster
3.8
4

Towards an Understanding of Graph Sequence Models

ICLR 2025Rejected

202410