MultistakeholderAI
Collection
Foundational multimodal corpus for provenance-preserving, multistakeholder AI deliberation research in digital policy and Internet governance. • 4 items • Updated
Arwen Policy LoRA
LoRA (Low-Rank Adaptation) adapter for the Arwen Policy Base model. This adapter fine-tunes the base model for evidence-grounded policy deliberation and stakeholder-aware synthesis on the Arwen Policy Corpus.
Base Model
soyames/arwen-policy-base (Qwen 3 configured for policy analysis via Ollama).
Intended Use
- Evidence-grounded policy synthesis
- Stakeholder-aware deliberation
- Multistakeholder Internet governance analysis
Limitations
- Task-specific adaptation; not a general-purpose model.
- Performance depends on the quality and coverage of the underlying corpus.
- Should not be used as an authoritative policy-making tool.
Links
- Base model: https://huggingface.co/soyames/arwen-policy-base
- Dataset: https://huggingface.co/datasets/soyames/arwen-policy-corpus
- Interactive demo: https://huggingface.co/spaces/soyames/arwen-policy
- Source code: https://github.com/soyames/arwen-policy
Citation
See CITATION.cff in the project repository.
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support