|
|
--- |
|
|
library_name: transformers |
|
|
license: cc-by-nc-4.0 |
|
|
datasets: |
|
|
- oumi-ai/oumi-c2d-d2c-subset |
|
|
- oumi-ai/oumi-synthetic-claims |
|
|
- oumi-ai/oumi-synthetic-document-claims |
|
|
language: |
|
|
- en |
|
|
base_model: |
|
|
- meta-llama/Llama-3.1-8B-Instruct |
|
|
--- |
|
|
|
|
|
[](https://github.com/oumi-ai/oumi) |
|
|
[](https://github.com/oumi-ai/oumi) |
|
|
|
|
|
[](https://oumi.ai/docs/en/latest/index.html) |
|
|
[](https://oumi.ai/blog) |
|
|
[](https://discord.gg/oumi) |
|
|
|
|
|
# oumi-ai/HallOumi-8B-classifier |
|
|
|
|
|
<!-- Provide a quick summary of what the model is/does. --> |
|
|
|
|
|
Introducing **HallOumi-8B-classifier**, a _fast_ **SOTA hallucination detection model**, outperforming DeepSeek R1, OpenAI o1, Google Gemini 1.5 Pro, and Anthropic Sonnet 3.5 at only 8 billion parameters! |
|
|
|
|
|
<!-- Give HallOumi a try now! --> |
|
|
|
|
|
<!-- * Demo: https://oumi.ai/halloumi-demo --> |
|
|
<!-- * Github: https://github.com/oumi-ai/oumi/tree/main/configs/projects/halloumi --> |
|
|
|
|
|
| Model | Balanced Accuracy | Macro F1 Score | Open Source? | Model Size | |
|
|
| --------------------- | ----------------- | --------------------------------------- | ------------ | ---------- | |
|
|
| **HallOumi-8B-classifier** | **76.8% ± 2.0%** | **78.5% ± 2.1%** | ✔️ | 8B | |
|
|
| Anthropic Sonnet 3.5 | 67.3% ± 2.7% | 69.6% ± 2.8% | ❌ | ?? | |
|
|
| OpenAI o1-preview | 64.5% ± 2.0% | 65.9% ± 2.3% | ❌ | ?? | |
|
|
| DeepSeek R1 | 60.7% ± 2.1% | 61.6% ± 2.5% | ✔️ | 671B | |
|
|
| Llama 3.1 405B | 58.7% ± 1.7% | 58.8% ± 2.4% | ✔️ | 405B | |
|
|
| Google Gemini 1.5 Pro | 52.9% ± 1.0% | 48.2% ± 1.8% | ❌ | ?? | |
|
|
|
|
|
Demo GIF: TODO |
|
|
|
|
|
|
|
|
**HallOumi-8B-classifier**, the hallucination classification model built with Oumi, is an end-to-end binary classification system that enables *fast and accurate* assessment of the hallucination probability of any written content (AI or human-generated). |
|
|
* ✔️ Fast with high accuracy |
|
|
* ✔️ Per-claim support (must call once per claim) |
|
|
|
|
|
|
|
|
## Hallucinations |
|
|
Hallucinations are often cited as the most important issue with being able to deploy generative models in numerous commercial and personal applications, and for good reason: |
|
|
|
|
|
* [Lawyers sanctioned for briefing where ChatGPT cited 6 fictitious cases](https://www.reuters.com/legal/new-york-lawyers-sanctioned-using-fake-chatgpt-cases-legal-brief-2023-06-22/) |
|
|
* [Air Canada required to honor refund policy made up by its AI support chatbot](https://www.wired.com/story/air-canada-chatbot-refund-policy/) |
|
|
* [AI suggesting users should make glue pizza and eat rocks](https://www.bbc.com/news/articles/cd11gzejgz4o) |
|
|
|
|
|
It ultimately comes down to an issue of **trust** — generative models are trained to produce outputs which are **probabilistically likely**, but not necessarily **true**. |
|
|
While such tools are certainly useful in the right hands, being unable to trust them prevents AI from being adopted more broadly, |
|
|
where it can be utilized safely and responsibly. |
|
|
|
|
|
## Building Trust with Verifiability |
|
|
To be able to begin trusting AI systems, we have to be able to verify their outputs. To verify, we specifically mean that we need to: |
|
|
|
|
|
* Understand the **truthfulness** of a particular statement produced by any model. |
|
|
* Understand what **information supports that statement’s truth** (or lack thereof) |
|
|
* Have **full traceability** connecting the statement to that information. |
|
|
|
|
|
Missing any one of these aspects results in a system that cannot be verified and therefore cannot be trusted; |
|
|
however, this is not enough, as we have to be capable of doing these things in a way that is **meticulous**, **scalable**, and **human-readable**. |
|
|
|
|
|
- **Developed by:** [Oumi AI](https://oumi.ai/) |
|
|
- **Model type:** Small Language Model |
|
|
- **Language(s) (NLP):** English |
|
|
- **License:** [CC-BY-NC-4.0](https://creativecommons.org/licenses/by-nc/4.0/deed.en) |
|
|
- **Finetuned from model:** [Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct) |
|
|
<!-- - **Demo:** [HallOumi Demo](https://oumi.ai/halloumi) --> |
|
|
|
|
|
--- |
|
|
|
|
|
## Uses |
|
|
|
|
|
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. --> |
|
|
Use to verify claims/detect hallucinations in scenarios where a known source of truth is available. |
|
|
|
|
|
<!-- Demo: https://oumi.ai/halloumi --> |
|
|
|
|
|
## Out-of-Scope Use |
|
|
|
|
|
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> |
|
|
Smaller LLMs have limited capabilities and should be used with caution. Avoid using this model for purposes outside of claim verification. |
|
|
|
|
|
## Bias, Risks, and Limitations |
|
|
|
|
|
<!-- This section is meant to convey both technical and sociotechnical limitations. --> |
|
|
This model was finetuned with Llama-3.1-405B-Instruct data on top of a Llama-3.1-8B-Instruct model, so any biases or risks associated with those models may be present. |
|
|
|
|
|
## Training Details |
|
|
|
|
|
### Training Data |
|
|
|
|
|
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> |
|
|
Training data: |
|
|
- [oumi-ai/oumi-synthetic-document-claims](https://huggingface.co/datasets/oumi-ai/oumi-synthetic-document-claims) |
|
|
- [oumi-ai/oumi-synthetic-claims](https://huggingface.co/datasets/oumi-ai/oumi-synthetic-claims) |
|
|
- [oumi-ai/oumi-anli-subset](https://huggingface.co/datasets/oumi-ai/oumi-anli-subset) |
|
|
- [oumi-ai/oumi-c2d-d2c-subset](https://huggingface.co/datasets/oumi-ai/oumi-c2d-d2c-subset) |
|
|
|
|
|
### Training Procedure |
|
|
|
|
|
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. --> |
|
|
Training notebook: Coming Soon |
|
|
|
|
|
## Evaluation |
|
|
|
|
|
<!-- This section describes the evaluation protocols and provides the results. --> |
|
|
Eval notebook: Coming Soon |
|
|
|
|
|
## Environmental Impact |
|
|
|
|
|
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> |
|
|
|
|
|
- **Hardware Type:** A100-80GB |
|
|
- **Hours used:** 1.5 (4 * 8 GPUs) |
|
|
- **Cloud Provider:** Google Cloud Platform |
|
|
- **Compute Region:** us-east5 |
|
|
- **Carbon Emitted:** 0.15 kg |
|
|
|
|
|
## Citation |
|
|
|
|
|
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> |
|
|
|
|
|
``` |
|
|
@misc{oumiHalloumi8BClassifier, |
|
|
author = {Achlioptas Panos, Jeremiah Greer, Aisopos Kostas, Schuler A. Michael, Elachqar Oussama, Koukoumidis Emmanouil}, |
|
|
title = {HallOumi-8B-classifier}, |
|
|
month = {March}, |
|
|
year = {2025}, |
|
|
url = {https://huggingface.co/oumi-ai/HallOumi-8B-classifier} |
|
|
} |
|
|
|
|
|
@software{oumi2025, |
|
|
author = {Oumi Community}, |
|
|
title = {Oumi: an Open, End-to-end Platform for Building Large Foundation Models}, |
|
|
month = {January}, |
|
|
year = {2025}, |
|
|
url = {https://github.com/oumi-ai/oumi} |
|
|
} |
|
|
``` |
