Transformers
PyTorch
English
language-model
graph-attention
adaptive-depth
temporal-decay
efficient-llm
Eval Results (legacy)
Instructions to use vigneshwar234/TemporalMesh-Transformer with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use vigneshwar234/TemporalMesh-Transformer with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("vigneshwar234/TemporalMesh-Transformer", dtype="auto") - Notebooks
- Google Colab
- Kaggle
v3 release: full paper, 5 TikZ diagrams, 25 tables, 5 algorithms β all open-source
#2
by vigneshwar234 - opened
TMT v3 Full Paper Released
The complete v3 paper is now available on Zenodo with:
- 20+ dense pages, NeurIPS-style two-column layout
- 5 TikZ architecture diagrams
- 25+ benchmark tables
- 5 formal algorithms (pseudocode)
- 4 real-world case studies (chatbot 100M users, legal 64K tokens, scientific papers, medical records)
- Full FLOP/memory derivations
- Quantisation robustness (only +3.7 PPL at INT4 vs +9.1 for vanilla)
- Cross-domain generalisation across 6 domains
DOI: 10.5281/zenodo.20287197
Paper: https://zenodo.org/records/20287390
Code: https://github.com/vignesh2027/TemporalMesh-Transformer
Demo: https://huggingface.co/spaces/vigneshwar234/TemporalMesh-Transformer-Demo