# Third-party notices

The repository's original code, prose, and site assets are released under the
[Unlicense](LICENSE). That dedication does not replace upstream licenses.

## SmolLM2-135M-Instruct

The adapter uses
[HuggingFaceTB/SmolLM2-135M-Instruct](https://huggingface.co/HuggingFaceTB/SmolLM2-135M-Instruct),
revision `12fd25f77366fa6b3b4b768ec3050bf629380bac`. The upstream model is
licensed under Apache License 2.0. The repository does not include the base
weights; the scripts download them from Hugging Face.

## Banking77

The experiment uses
[Banking77](https://github.com/PolyAI-LDN/task-specific-datasets/tree/57ec275d8078af65b7731c2a98be812d844a6d6b/banking_data),
revision `57ec275d8078af65b7731c2a98be812d844a6d6b`. PolyAI distributes the
dataset under Creative Commons Attribution 4.0 International.

Please cite Iñigo Casanueva, Tadas Temčinas, Daniela Gerz, Matthew Henderson,
and Ivan Vulić, “Efficient Intent Detection with Dual Sentence Encoders,” 2020.

The committed `adapter-banking77/` artifact and Banking77-derived evidence are
distributed under CC BY 4.0. No endorsement by PolyAI or Hugging Face is implied.