nokia-applied-research/AnyJev
Turn any LLM into a Jev-style decision model: typed decisions, real probabilities, no training. (continue updating)
From the publisher
README & documentation
Source previewRead the project’s overview, installation instructions and usage examples. The original README is the source of truth.
[](https://pypi.org/project/anyjev/) [](https://pypi.org/project/anyjev/) [](https://github.com/nokia-applied-research/AnyJev/actions/workflows/ci.yml) [](LICENSE) English · 简体中文 · 🚀 Usage · 📊 Results · 🧭 Roadmap · 📖 Levels contract Jiamu Zhang1 Tianze Yang1 Yucheng Shi2 Liang Wu1 1 Nokia, Sunnyvale, CA 2 Tencent Hunyuan Qwen3-8B on a real BANKING77 item. Every number is a model output. Ask any open LLM a typed question and get back a decision with a probability you can threshold, read from one prefill of its next-token distribution. No generation, no parsing, no fine-tuning. Raw logits change their answer when you reorder the options, and their confidence cannot be trusted; AnyJev fixes the first with zero labels and the second with a few hundred. Qwen3-8B, BANKING77 20-way, 300 test items. Full table incl. every ablation: docs/resultsbench.md The last row is the point. Accuracy moves by 6 points, but the share of traffic you can safely automate goes from 7.7% to 52.0%, a 6.8×…
Read the full README ↗ · Preview checked 2026-09-23T12:47:25.492Z
Inside the original README — Document outline
- ✨ What it does
- 🚀 Usage
- 🧠 How it works
- 🔁 A head that maintains itself
- 📊 Results
- 🧭 Roadmap
- 🔍 Limitations
- 🤝 Contributing and citation
Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.
Links from the README
References supplied by the publisher, not independently verified endorsements. Check the destination before downloading files or entering credentials.
Documentation belongs to its respective authors. Reported project/model license: Apache-2.0. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.
What this repository does
Turn any LLM into a Jev-style decision model: typed decisions, real probabilities, no training. (continue updating)
Repository facts
- Owner
- nokia-applied-research
- Primary language
- Python
- Stars
- 285
- Forks
- 41
- Open issues + pull requests
- 2
- License
- Apache-2.0
- Archived
- No
- Default branch
- main
- Created
- 2026-09-21T05:55:40.000Z
- Last push
- 2026-09-23T08:51:07.000Z
Topics and intended use
Owner-supplied topics: calibration, decision-model, jev, jev-model, llm, system-one, transformers, vllm
Review the README for scope, installation, examples and limitations. We do not run repository code or certify it.
Evaluate before installing
Review licensing and dependencies, inspect recent commits and unresolved issues, and test in an isolated environment before production use. Stars and forks alone cannot answer these questions.
Source and freshness
Source: GitHub. Metadata observed 2026-09-23T12:03:29.066Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.
Related repositories
EbookFoundation/free-programming-books
donnemartin/system-design-primer