Eco-Transformers: Carbon-Efficient and Accelerated Inference through Self-Pruning and Activation Freezing in Large Language Models

The question behind it

What can a model see that a person cannot?

Projects I have run with undergraduate teams: infection detection in mushroom farms, solar output under cloud cover, fake review and fake news detection, gesture recognition, text summarisation, and carbon-efficient inference in large language models.

A plain-language summary of this paper is on its way here. The full text will follow wherever the publisher permits me to host it.

Cite this paper

@inproceedings{shriaarthy2025eco,
  title   = {Eco-Transformers: Carbon-Efficient and Accelerated Inference through Self-Pruning and Activation Freezing in Large Language Models},
  author  = {E Shriaarthy and S Sangeetha and RS Niranjana and R Rohini and B Sivanesan and et al.},
  booktitle = {2025 ICNGCS, pp. 1-8},
  year    = {2025},
  doi     = {10.1109/icngcs64900.2025.11183217},
  url     = {https://doi.org/10.1109/icngcs64900.2025.11183217},
}

Related work in this thread