PaperHub

Zhao Song

~Zhao_Song3

52
论文总数
26.0
年均投稿
5.4
平均评分
接收情况26/52
会议分布
ICLR
34
NeurIPS
11
ICML
7

发表论文 (52 篇)

202536

7.8
4

Differential Privacy for Euclidean Jordan Algebra with Applications to Private Symmetric Cone Programming

NeurIPS 2025Poster
4.7
3

Revisiting Quantum Algorithms for Linear Regressions: Quadratic Speedups without Data-Dependent Parameters

ICLR 2025Rejected
3.5
4

Towards Sampling Data Structures for Tensor Products

ICLR 2025Rejected
5.8
4

Efficient Alternating Minimization with Applications to Weighted Low Rank Approximation

ICLR 2025Poster
3.5
4

Unlocking the Theory Behind Scaling 1-Bit Neural Networks

ICLR 2025withdrawn
6.8
4

Faster Algorithms for Structured Linear and Kernel Support Vector Machines

ICLR 2025Poster
4.8
5

Fast RoPE Attention: Combining the Polynomial Method and Fast Fourier Transform

ICLR 2025Rejected
4.0
3

Binary Hypothesis Testing for Softmax Models and Leverage Score Models

ICML 2025Poster
7.0
3

Deterministic Sparse Fourier Transform for Continuous Signals with Frequency Gap

ICML 2025Poster
3.8
4

An $O(k\log n)$ Time Fourier Set Query Algorithm

ICLR 2025Rejected
5.8
4

A Dynamic Low-Rank Fast Gaussian Transform

ICLR 2025Rejected
4.8
4

Training Overparametrized Neural Networks in Sublinear Time

ICLR 2025Rejected
7.2
4

Discrepancy Minimization in Input-Sparsity Time

ICML 2025Spotlight
3.0
4

Binary Hypothesis Testing for Softmax Models and Leverage Score Models

ICLR 2025Rejected
4.5
4

Dynamic Kernel Sparsifiers

ICLR 2025Rejected
6.8
4

Faster Algorithms for Structured John Ellipsoid Computation

NeurIPS 2025Poster
6.4
4

Attention Mechanism, Max-Affine Partition, and Universal Approximation

NeurIPS 2025Poster
5.0
4

Towards Infinite-Long Prefix in Transformer

ICLR 2025Rejected
2.3
3

One Pass Streaming Algorithm for Super Long Token Attention Approximation in Sublinear Space

ICLR 2025Rejected
5.5
4

Tensor Attention Training: Provably Efficient Learning of Higher-order Transformers

ICLR 2025Rejected
7.3
4

Efficient $k$-Sparse Band–Limited Interpolation with Improved Approximation Ratio

NeurIPS 2025Poster
5.8
4

Differential Privacy of Cross-Attention with Provable Guarantee

ICLR 2025Rejected
4.3
4

On Differentially Private String Distances

ICLR 2025withdrawn
6.3
3

Fine-grained Attention I/O Complexity: Comprehensive Analysis for Backward Passes

ICLR 2025Rejected
8.3
4

On Differential Privacy for Adaptively Solving Search Problems via Sketching

ICML 2025Oral
5.3
4

Pretrained Transformers are Deep Optimizers: Provable In-Context Learning for Deep Model Training

ICLR 2025Rejected
6.1
4

In-Context Deep Learning via Transformer Models

ICML 2025Poster
5.8
4

Computational Limits of Low-Rank Adaptation (LoRA) Fine-Tuning for Transformer Models

ICLR 2025Poster
4.7
3

Conv-Basis: A New Paradigm for Efficient Attention Inference and Gradient Computation in Transformers

ICLR 2025Rejected
5.8
4

Multi-Layer Transformers Gradient Can be Approximated in Almost Linear Time

ICLR 2025Rejected
5.8
4

Beyond Linear Approximations: A Novel Pruning Approach for Attention Matrix

ICLR 2025Poster
3.0
4

HSR-Enhanced Sparse Attention Acceleration

ICLR 2025withdrawn
6.3
4

Fundamental Limits of Prompt Tuning Transformers: Universality, Capacity and Efficiency

ICLR 2025Poster
4.8
3

Fundamental Limits of Visual Autoregressive Transformers: Universal Approximation Abilities

ICML 2025Poster
6.3
3

Dissecting Submission Limit in Desk-Rejections: A Mathematical Analysis of Fairness in AI Conference Policies

ICML 2025Poster
6.4
5

High-Order Flow Matching: Unified Framework and Sharp Statistical Rates

NeurIPS 2025Poster

202416

4.5
4

On Socially Fair Regression and Low-Rank Approximation

ICLR 2024Rejected
6.3
4

On Socially Fair Low-Rank Approximation and Column Subset Selection

NeurIPS 2024Poster
5.8
4

The Fine-Grained Complexity of Gradient Computation for Training Large Language Models

NeurIPS 2024Poster
3.0
4

Unmasking Transformers: A Theoretical Approach to Data Recovery via Attention Weights

ICLR 2024withdrawn
6.0
3

The Closeness of In-Context Learning and Weight Shifting for Softmax Regression

NeurIPS 2024Poster
3.7
3

Fine-tune Language Models to Approximate Unbiased In-context Learning

ICLR 2024Rejected
5.0
3

The Closeness of In-Context Learning and Weight Shifting for Softmax Regression

ICLR 2024Rejected
8.0
3

How to Capture Higher-order Correlations? Generalizing Matrix Softmax Attention to Kronecker Computation

ICLR 2024Spotlight
5.0
4

Algorithm and Hardness for Dynamic Attention Maintenance in Large Language Models

ICLR 2024Rejected
6.0
3

Low Rank Matrix Completion via Robust Alternating Minimization in Nearly Linear Time

ICLR 2024Poster
5.8
4

A Sublinear Adversarial Training Algorithm

ICLR 2024Poster
3.7
3

Parameter-Efficient Fine-Tuning via Partially Decomposable Loss Analysis and Sharing

ICLR 2024withdrawn
-

On the Similarity between Attention and SVM on the Token Separation and Selection Behavior

ICLR 2024withdrawn
7.3
3

Log-concave Sampling from a Convex Body with a Barrier: a Robust and Unified Dikin Walk

NeurIPS 2024Poster
6.3
4

On Statistical Rates and Provably Efficient Criteria of Latent Diffusion Transformers (DiTs)

NeurIPS 2024Poster
7.0
3

Metric Transforms and Low Rank Representations of Kernels for Fast Attention

NeurIPS 2024Spotlight