Yifan Zhang

deep-delta-learning. Official Project Page for Deep Delta Learning (https://arxiv.org/abs/2601.00417)

HLA. Official Project Page for HLA: Higher-order Linear Attention (https://arxiv.org/abs/2510.27258)

AutoMathText. [ACL 2025 Findings] Autonomous Data Selection with Zero-shot Generative Classifiers for Mathematical Texts (https://huggingface.co/papers/2402.07625)

Matrix-SSL. [ICML 2024] Matrix Information Theory for Self-supervised Learning (https://arxiv.org/abs/2305.17326)

FALCON. Official Project Page for FALCON: Fast-Weight Attention for Continual Learning (https://yifanzhang-pro.github.io/FALCON)

Kernel-InfoNCE. [ICLR 2024] Contrastive Learning Is Spectral Clustering On Similarity Graph (https://arxiv.org/abs/2303.15103)

MEA. Official Project Page for MEA: Matrix Exponential Attention

quantum-lattice. Official Project Page for "Exact Coset Sampling for Quantum Lattice Algorithms" (https://arxiv.org/abs/2509.12341)

lanser-cli. [Lanser-CLI] Official Implementation of "Reinforcement Learning from Compiler and Language Server Feedback" (https://arxiv.org/abs/2510.22907)

M-MAE. [ICML 2024] Matrix Variational Masked Autoencoder (M-MAE) for ICML paper "Information Flow in Self-Supervised Learning" (https://arxiv.org/abs/2309.17281)

how-to-train-frontier-models.

BlueMO. BlueMO: A Comprehensive Collection of Challenging Mathematical Olympiad Problems from the Little Blue Book Series

StackMathQA. StackMathQA: A Curated Collection of 2 Million Mathematical Questions and Answers Sourced from Stack Exchange

Revisiting-SVRPG-LLM-RL. HTML

ICS. All 8 Labs of Course: Introduction to Computer Systems with textbook CSAPP

residual-stream-duality. Official Project Page for Residual Stream Duality in Modern Transformer Architectures (https://arxiv.org/abs/2603.16039)

RelationMatch. RelationMatch: Matching In-batch Relationships for Semi-supervised Learning (https://arxiv.org/abs/2305.10397)

fast-weight-attention. Official Project Page for Falcon: Fast Weight Attention for Continual Learning

syntax-semantics. TemplateMath: Syntactic Data Generation for Mathematical Problems

Matrix-LLM. Official Implementation of ICML 2024 paper 'Matrix Cross-Entropy for Large Language Models' (https://arxiv.org/abs/2305.17326)

lagrange-superintelligence. Lagrange Points Are the Next Frontier of Superintelligence

Rethinking-SWA. Official Project Page for "Rethinking SWA": Why Short Sliding Window Attention Will Replace ShortConv in Modern Architectures

deep-rotary-learning. Deep Rotary Learning

matrix-hidden-states. Matrix Hidden States

ShortSWA-Ngram-Embedding. Short Sliding Window Attention (ShortSWA) is the next-generation N-gram embedding.

yifanzhang-pro. Yifan Zhang

TPA. [NeurIPS 2025 Spotlight] TPA: Tensor ProducT ATTenTion Transformer (T6) (https://arxiv.org/abs/2501.06425)

efficient-consistent-explanations. [NeurIPS 2023] Trade-off Between Efficiency and Consistency for Removal-based Explanations (https://arxiv.org/abs/2210.17426)

RPG. Official implementation of Regularized Policy Gradient (RPG) (https://arxiv.org/abs/2505.17508)

general-preference-model. [ICML 2025] Beyond Bradley-Terry Models: A General Preference Model for Language Model Alignment (https://arxiv.org/abs/2410.02197)

HLA. Higher Attention I: Higher-Order Linear Attention with Parallel Scans

quantum-lattice.

lm-theory. A Markov Categorical Framework for Language Modeling

AutoMathText-2.1.