Beyond Multiple Choice: Evaluating Steering Vectors for Adaptive Free-Form Summarization

This repository contains the implementation and experimental results for the paper "Beyond Multiple Choice: Evaluating Steering Vectors for Adaptive Free-Form Summarization".

Paper

This work has been published at the ICML 2025 Workshop on Reliable and Responsible Foundation Models.

• ArXiv: https://arxiv.org/abs/2505.24859
• OpenReview: https://openreview.net/forum?id=sbm53EmmGp
• PDF: icml_2025_r2fm_workshop.pdf (included in this repository)

Project Structure

The repository is organized as follows:

.
├── config/             # Configuration files for experiments
├── data/               # Results from the experiments
├── datasets/           # Datasets used in the experiments
│   ├── sentiment/      # Sentiment vectors synthetic training data
│   ├── readability/    # Readability vectors synthetic training data
│   ├── toxicity/       # Toxicity vectors synthetic training data
│   └── topic/          # Topic vectors training representations
├── notebooks/          # Jupyter notebooks for exploratory analysis and visualization
├── scripts/            # Supporting scripts for data preparation and preprocessing
├── src/                # Core source code
│   ├── __init__.py
│   ├── experiments/    # Main steering experiments
│   ├── utils/
│   ├── plot_results.py # plot results for token reweighting
│   ├── run_experiments.py # run token reweighting experiments
│   ├── score_results.py   # score token reweighting experiments
├── tests/              # Tests for some of the steering functionalities
├── pyproject.toml      # Poetry configuration and dependencies
├── poetry.lock         # Locked dependencies
├── LICENSE             # License for the repository
├── README.md           # Project documentation (this file)

Installation

This project uses Python 3.12 and Poetry for dependency management. Follow these steps to set up the environment:

1. Clone the repository:

   git clone [repository-url]
   cd [repository-name]

2. Install dependencies using Poetry:

   poetry install

3. Activate the virtual environment:

   poetry shell

Development

The project includes development dependencies for testing (pytest), type checking (mypy), code formatting (black), and linting (ruff), as specified in pyproject.toml.

Datasets

The synthetic datasets used for training the steering vectors can be found in the following locations:

• Sentiment data: datasets/sentiment/
• Readability data: datasets/readability/
• Toxicity data: datasets/toxicity/
• Topic representations: datasets/topic/

License

This project is licensed under the MIT License. See the LICENSE file for details.