spbkuznetsov/few-shot-multimodal-notebook

Few Shot Multimodal — open questions

Overview

This is an exploratory note for Few Shot Multimodal. It records the intended comparison, likely confounders, and reproducibility requirements before any benchmark result is reported.

Feedback on the structure is welcome; the code is intentionally short.

Contact

Found a mistake? Open an issue — patches and pointers both welcome.

Related work

1. survey-level references come first
2. benchmark papers follow
3. negative results are noted where they exist

Open questions

1. which baselines are actually comparable under matched capacity?
2. what is the minimum data exposure for a meaningful signal?
3. which confounders would invalidate a naive comparison?

Scope and limitations

The note is intentionally exploratory. It does not claim benchmark improvements, completed ablations, released code, or a trained checkpoint. References and proposed datasets provide a starting point for verification rather than evidence that the study has already been run.

What is covered

1. research question scope and likely confounders
2. one proposed comparison against matched baselines
3. evaluation context such as task-appropriate public benchmarks named in the main note
4. reproducibility checks and open questions

How to read this repository

Start with review.md for the full note. Sections labeled as plans or hypotheses should not be interpreted as experimental results. If results are added later, they should include dataset versions, commands, seeds, hardware, and raw logs.

Files

1. review.md — primary artifact
2. README.md — this documentation
3. config.json — architecture configuration
4. training_args.json — default experiment settings
5. model.safetensors — initialization checkpoint

License

Distributed under the cc0-1.0 license. External datasets keep their own terms — check them separately.