CUHK SphereLab

university

https://spherelab.ai/

AI & ML interests

None defined yet.

Recent Activity

sihany updated a collection about 22 hours ago

sihany updated a collection about 23 hours ago

sihany updated a collection about 23 hours ago

View all activity

Papers

PEFT-Arena: Understanding Parameter-Efficient Finetuning from a Stability-Plasticity Perspective

Model Merging with Functional Dual Anchors

View all Papers

updated a collection about 22 hours ago

TTR-Compressor

7 items • Updated about 22 hours ago

updated a collection about 23 hours ago

TTR-Compressor

7 items • Updated about 22 hours ago

updated a collection 2 days ago

TTR-Compressor

7 items • Updated about 22 hours ago

submitted a paper to Daily Papers 19 days ago

What Makes Recurrence Effective in Looped Language Models?

Paper • 2609.36636 • Published 21 days ago • 11

authored a paper 4 months ago

Sample-Efficient Post-Training for LEGO Spatial-Physics Reasoning

Paper • 2606.07602 • Published May 29 • 6

authored a paper 5 months ago

PEFT-Arena: Understanding Parameter-Efficient Finetuning from a Stability-Plasticity Perspective

Paper • 2605.28819 • Published May 27 • 8

authored a paper 5 months ago

PEFT-Arena: Understanding Parameter-Efficient Finetuning from a Stability-Plasticity Perspective

Paper • 2605.28819 • Published May 27 • 8

updated a collection 5 months ago

PEFT-Arena

PEFT-Arena: Understanding Parameter-Efficient Finetuning from a Stability-Plasticity Perspective • 4 items • Updated May 28 • 1

submitted a paper to Daily Papers 5 months ago

PEFT-Arena: Understanding Parameter-Efficient Finetuning from a Stability-Plasticity Perspective

Paper • 2605.28819 • Published May 27 • 8

submitted a paper to Daily Papers 5 months ago

Pion: A Spectrum-Preserving Optimizer via Orthogonal Equivalence Transformation

Paper • 2605.12492 • Published May 12 • 6

updated a dataset 6 months ago

SphereLab/PEFTArena-data

Viewer • Updated Apr 25 • 73.7k • 13

published a dataset 6 months ago

SphereLab/PEFTArena-data

Viewer • Updated Apr 25 • 73.7k • 13

authored a paper 6 months ago

Stabilizing Rubric Integration Training via Decoupled Advantage Normalization

Paper • 2603.26535 • Published Mar 27 • 4

authored a paper 8 months ago

PyVision-RL: Forging Open Agentic Vision Models via RL

Paper • 2602.20739 • Published Feb 24 • 32