Jade's picture

Jade

euclaise

·

AI & ML interests

None yet

Recent Activity

upvoted a paper 1 day ago

Self-Taught Self-Correction for Small Language Models

upvoted a paper 1 day ago

Shifting Long-Context LLMs Research from Input to Output

upvoted a paper 1 day ago

Transformers without Normalization

View all activity

Organizations

euclaise's activity

upvoted 3 papers 1 day ago

Self-Taught Self-Correction for Small Language Models

Paper • 2503.08681 • Published 7 days ago • 12

Shifting Long-Context LLMs Research from Input to Output

Paper • 2503.04723 • Published 12 days ago • 19

Transformers without Normalization

Paper • 2503.10622 • Published 5 days ago • 122

upvoted 2 papers 6 days ago

Optimizing Test-Time Compute via Meta Reinforcement Fine-Tuning

Paper • 2503.07572 • Published 8 days ago • 36

LMM-R1: Empowering 3B LMMs with Strong Reasoning Abilities Through Two-Stage Rule-Based RL

Paper • 2503.07536 • Published 8 days ago • 77

upvoted 4 papers 7 days ago

DiffCLIP: Differential Attention Meets CLIP

Paper • 2503.06626 • Published 10 days ago • 4

Linear-MoE: Linear Sequence Modeling Meets Mixture-of-Experts

Paper • 2503.05447 • Published 11 days ago • 7

Forgetting Transformer: Softmax Attention with a Forget Gate

Paper • 2503.02130 • Published 15 days ago • 27

RePO: ReLU-based Preference Optimization

Paper • 2503.07426 • Published 8 days ago • 1

upvoted a paper 11 days ago

HybridNorm: Towards Stable and Efficient Transformer Training via Hybrid Normalization

Paper • 2503.04598 • Published 12 days ago • 17

upvoted 3 papers 12 days ago

PLDR-LLMs Learn A Generalizable Tensor Operator That Can Replace Its Own Deep Neural Net At Inference

Paper • 2502.13502 • Published 28 days ago • 3

Liger: Linearizing Large Language Models to Gated Recurrent Structures

Paper • 2503.01496 • Published 16 days ago • 15

Cognitive Behaviors that Enable Self-Improving Reasoners, or, Four Habits of Highly Effective STaRs

Paper • 2503.01307 • Published 16 days ago • 31

upvoted a paper 16 days ago

NeoBERT: A Next-Generation BERT

Paper • 2502.19587 • Published 20 days ago • 38

upvoted a paper 20 days ago

Scale-Distribution Decoupling: Enabling Stable and Effective Training of Large Language Models

Paper • 2502.15499 • Published 25 days ago • 13

upvoted a paper 21 days ago

Stable-SPAM: How to Train in 4-Bit More Stably than 16-Bit Adam

Paper • 2502.17055 • Published 23 days ago • 16

upvoted a paper 23 days ago

AlphaMaze: Enhancing Large Language Models' Spatial Intelligence via GRPO

Paper • 2502.14669 • Published 26 days ago • 12

upvoted 3 papers 24 days ago

Revisiting the Test-Time Scaling of o1-like Models: Do they Truly Possess Test-Time Scaling Capabilities?

Paper • 2502.12215 • Published 30 days ago • 16

You Do Not Fully Utilize Transformer's Representation Capacity

Paper • 2502.09245 • Published Feb 13 • 34

Cramming 1568 Tokens into a Single Vector and Back Again: Exploring the Limits of Embedding Space Capacity

Paper • 2502.13063 • Published 28 days ago • 67