Fromthesky Research Labs

company

https://fromtheskyresearchlabs.com/

fromtheskylabs

fromthesky-research-labs

Activity Feed

AI & ML interests

Large Language Models, Transformers, Natural Language Processing

Recent Activity

fromthesky authored a paper about 7 hours ago

Training and Inference Dynamics of PLDR-LLMs: Row-Map Collapse, Renormalization, and Predictive Reduction

fromthesky submitted a paper about 10 hours ago

Training and Inference Dynamics of PLDR-LLMs: Row-Map Collapse, Renormalization, and Predictive Reduction

fromthesky updated a collection about 2 months ago

PLDR-LLMs with KVG cache (Pytorch/Transformers)

View all activity

Papers

Training and Inference Dynamics of PLDR-LLMs: Row-Map Collapse, Renormalization, and Predictive Reduction

Power law graph attention: exact generalization of scaled dot-product attention, empirical collapse at inference

View all Papers

FromtheskyResearchLabs 's papers 7

Submitted by

Burc Gokden

Training and Inference Dynamics of PLDR-LLMs: Row-Map Collapse, Renormalization, and Predictive Reduction

Fromthesky Research Labs

0 1

Submitted by

Burc Gokden

Power law graph attention: exact generalization of scaled dot-product attention, empirical collapse at inference

Fromthesky Research Labs

1 2

Submitted by

Burc Gokden

PLDR-LLMs Reason At Self-Organized Criticality

Fromthesky Research Labs

1 2

Submitted by

Burc Gokden

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

Fromthesky Research Labs

0 2

AI & ML interests

Recent Activity

Papers

Team members 1

Training and Inference Dynamics of PLDR-LLMs: Row-Map Collapse, Renormalization, and Predictive Reduction

Power law graph attention: exact generalization of scaled dot-product attention, empirical collapse at inference

PLDR-LLMs Reason At Self-Organized Criticality

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

PLDR-LLM: Large Language Model from Power Law Decoder Representations

Power Law Graph Transformer for Machine Translation and Representation Learning

CoulGAT: An Experiment on Interpretability of Graph Attention Networks