https://github.com/cvs-health/uqlm
UQLM (Uncertainty Quantification for Language Models) is a Python package for UQ-based LLM hallucination detection
https://github.com/cvs-health/uqlm
ai-evaluation ai-safety confidence-estimation confidence-score hallucination hallucination-detection hallucination-evaluation hallucination-mitigation llm llm-evaluation llm-hallucination llm-safety uncertainty-estimation uncertainty-quantification
Last synced: 30 days ago
JSON representation
UQLM (Uncertainty Quantification for Language Models) is a Python package for UQ-based LLM hallucination detection
- Host: GitHub
- URL: https://github.com/cvs-health/uqlm
- Owner: cvs-health
- License: apache-2.0
- Created: 2025-04-17T23:12:21.000Z (about 2 months ago)
- Default Branch: main
- Last Pushed: 2025-05-06T15:08:42.000Z (about 1 month ago)
- Last Synced: 2025-05-06T15:59:40.705Z (about 1 month ago)
- Topics: ai-evaluation, ai-safety, confidence-estimation, confidence-score, hallucination, hallucination-detection, hallucination-evaluation, hallucination-mitigation, llm, llm-evaluation, llm-hallucination, llm-safety, uncertainty-estimation, uncertainty-quantification
- Language: Python
- Homepage: https://cvs-health.github.io/uqlm/latest/index.html
- Size: 11.6 MB
- Stars: 14
- Watchers: 0
- Forks: 7
- Open Issues: 5
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
- trackawesomelist - UQLM (⭐33) - of-the-art uncertainty quantification techniques  (Recently Updated / [May 10, 2025](/content/2025/05/10/README.md))
- awesome-langchain - UQLM - of-the-art uncertainty quantification techniques  (Tools / Services)
README
# uqlm: Uncertainty Quantification for Language Models
[](https://opensource.org/licenses/Apache-2.0)
[](https://github.com/psf/black)
[](https://github.com/cvs-health/uqlm/actions)
[](https://arxiv.org/abs/2504.19254)
UQLM is a Python library for Large Language Model (LLM) hallucination detection using state-of-the-art uncertainty quantification techniques.
## Installation
The latest version can be installed from PyPI:
```bash
pip install uqlm
```
## Hallucination Detection
UQLM provides a suite of response-level scorers for quantifying the uncertainty of Large Language Model (LLM) outputs. Each scorer returns a confidence score between 0 and 1, where higher scores indicate a lower likelihood of errors or hallucinations. We categorize these scorers into four main types:
| Scorer Type | Added Latency | Added Cost | Compatibility | Off-the-Shelf / Effort |
|------------------------|----------------------------------------------------|------------------------------------------|-----------------------------------------------------------|---------------------------------------------------------|
| [Black-Box Scorers](#black-box-scorers-consistency-based) | ⏱️ Medium-High (multiple generations & comparisons) | 💸 High (multiple LLM calls) | 🌍 Universal (works with any LLM) | ✅ Off-the-shelf |
| [White-Box Scorers](#white-box-scorers-token-probability-based) | ⚡ Minimal (token probabilities already returned) | ✔️ None (no extra LLM calls) | 🔒 Limited (requires access to token probabilities) | ✅ Off-the-shelf |
| [LLM-as-a-Judge Scorers](#llm-as-a-judge-scorers) | ⏳ Low-Medium (additional judge calls add latency) | 💵 Low-High (depends on number of judges)| 🌍 Universal (any LLM can serve as judge) |✅ Off-the-shelf |
| [Ensemble Scorers](#ensemble-scorers) | 🔀 Flexible (combines various scorers) | 🔀 Flexible (combines various scorers) | 🔀 Flexible (combines various scorers) | ✅ Off-the-shelf (beginner-friendly); 🛠️ Can be tuned (best for advanced users) |
Below we provide illustrative code snippets and details about available scorers for each type.
### Black-Box Scorers (Consistency-Based)
These scorers assess uncertainty by measuring the consistency of multiple responses generated from the same prompt. They are compatible with any LLM, intuitive to use, and don't require access to internal model states or token probabilities.
**Example Usage:**
Below is a sample of code illustrating how to use the `BlackBoxUQ` class to conduct hallucination detection.
```python
from langchain_google_vertexai import ChatVertexAI
llm = ChatVertexAI(model='gemini-pro')
from uqlm import BlackBoxUQ
bbuq = BlackBoxUQ(llm=llm, scorers=["semantic_negentropy"], use_best=True)
results = await bbuq.generate_and_score(prompts=prompts, num_responses=5)
results.to_df()
```
Above, `use_best=True` implements mitigation so that the uncertainty-minimized responses is selected. Note that although we use `ChatVertexAI` in this example, any [LangChain Chat Model](https://js.langchain.com/docs/integrations/chat/) may be used. For a more detailed demo, refer to our [Black-Box UQ Demo](./examples/black_box_demo.ipynb).
**Available Scorers:**
* Non-Contradiction Probability ([Chen & Mueller, 2023](https://arxiv.org/abs/2308.16175); [Lin et al., 2025](https://arxiv.org/abs/2305.19187); [Manakul et al., 2023](https://arxiv.org/abs/2303.08896))
* Semantic Entropy ([Farquhar et al., 2024](https://www.nature.com/articles/s41586-024-07421-0); [Kuhn et al., 2023](https://arxiv.org/abs/2302.09664))
* Exact Match ([Cole et al., 2023](https://arxiv.org/abs/2305.14613); [Chen & Mueller, 2023](https://arxiv.org/abs/2308.16175))
* BERT-score ([Manakul et al., 2023](https://arxiv.org/abs/2303.08896); [Zheng et al., 2020](https://arxiv.org/abs/1904.09675))
* BLUERT-score ([Sellam et al., 2020](https://arxiv.org/abs/2004.04696))
* Cosine Similarity ([Shorinwa et al., 2024](https://arxiv.org/abs/2412.05563); [HuggingFace](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2))
### White-Box Scorers (Token-Probability-Based)
These scorers leverage token probabilities to estimate uncertainty. They are significantly faster and cheaper than black-box methods, but require access to the LLM's internal probabilities, meaning they are not necessarily compatible with all LLMs/APIs.
**Example Usage:**
Below is a sample of code illustrating how to use the `WhiteBoxUQ` class to conduct hallucination detection.
```python
from langchain_google_vertexai import ChatVertexAI
llm = ChatVertexAI(model='gemini-pro')
from uqlm import WhiteBoxUQ
wbuq = WhiteBoxUQ(llm=llm, scorers=["min_probability"])
results = await wbuq.generate_and_score(prompts=prompts)
results.to_df()
```
Again, any [LangChain Chat Model](https://js.langchain.com/docs/integrations/chat/) may be used in place of `ChatVertexAI`. For a more detailed demo, refer to our [White-Box UQ Demo](./examples/white_box_demo.ipynb).
**Available Scorers:**
* Minimum token probability ([Manakul et al., 2023](https://arxiv.org/abs/2303.08896))
* Length-Normalized Joint Token Probability ([Malinin & Gales, 2021](https://arxiv.org/abs/2002.07650))
### LLM-as-a-Judge Scorers
These scorers use one or more LLMs to evaluate the reliability of the original LLM's response. They offer high customizability through prompt engineering and the choice of judge LLM(s).
**Example Usage:**
Below is a sample of code illustrating how to use the `LLMPanel` class to conduct hallucination detection using a panel of LLM judges.
```python
from langchain_google_vertexai import ChatVertexAI
llm1 = ChatVertexAI(model='gemini-1.0-pro')
llm2 = ChatVertexAI(model='gemini-1.5-flash-001')
llm3 = ChatVertexAI(model='gemini-1.5-pro-001')
from uqlm import LLMPanel
panel = LLMPanel(llm=llm1, judges=[llm1, llm2, llm3])
results = await panel.generate_and_score(prompts=prompts)
results.to_df()
```
Note that although we use `ChatVertexAI` in this example, we can use any [LangChain Chat Model](https://js.langchain.com/docs/integrations/chat/) as judges. For a more detailed demo illustrating how to customize a panel of LLM judges, refer to our [LLM-as-a-Judge Demo](./examples/judges_demo.ipynb).
**Available Scorers:**
* Categorical LLM-as-a-Judge ([Manakul et al., 2023](https://arxiv.org/abs/2303.08896); [Chen & Mueller, 2023](https://arxiv.org/abs/2308.16175); [Luo et al., 2023](https://arxiv.org/abs/2303.15621))
* Continuous LLM-as-a-Judge ([Xiong et al., 2024](https://arxiv.org/abs/2306.13063))
* Panel of LLM Judges ([Verga et al., 2024](https://arxiv.org/abs/2404.18796))
### Ensemble Scorers
These scorers leverage a weighted average of multiple individual scorers to provide a more robust uncertainty/confidence estimate. They offer high flexibility and customizability, allowing you to tailor the ensemble to specific use cases.
**Example Usage:**
Below is a sample of code illustrating how to use the `UQEnsemble` class to conduct hallucination detection.
```python
from langchain_google_vertexai import ChatVertexAI
llm = ChatVertexAI(model='gemini-pro')
from uqlm import UQEnsemble
## ---Option 1: Off-the-Shelf Ensemble---
# uqe = UQEnsemble(llm=llm)
# results = await uqe.generate_and_score(prompts=prompts, num_responses=5)
## ---Option 2: Tuned Ensemble---
scorers = [ # specify which scorers to include
"exact_match", "noncontradiction", # black-box scorers
"min_probability", # white-box scorer
llm # use same LLM as a judge
]
uqe = UQEnsemble(llm=llm, scorers=scorers)
# Tune on tuning prompts with provided ground truth answers
tune_results = await uqe.tune(
prompts=tuning_prompts, ground_truth_answers=ground_truth_answers
)
# ensemble is now tuned - generate responses on new prompts
results = await uqe.generate_and_score(prompts=prompts)
results.to_df()
```
As with the other examples, any [LangChain Chat Model](https://js.langchain.com/docs/integrations/chat/) may be used in place of `ChatVertexAI`. For more detailed demos, refer to our [Off-the-Shelf Ensemble Demo](./examples/ensemble_off_the_shelf_demo.ipynb) (quick start) or our [Ensemble Tuning Demo](./examples/ensemble_tuning_demo.ipynb) (advanced).
**Available Scorers:**
* BS Detector ([Chen & Mueller, 2023](https://arxiv.org/abs/2308.16175))
* Generalized UQ Ensemble ([Bouchard & Chauhan, 2025](https://arxiv.org/abs/2504.19254))
## Documentation
Check out our [documentation site](https://cvs-health.github.io/uqlm/latest/index.html) for detailed instructions on using this package, including API reference and more.
## Example notebooks
Explore the following demo notebooks to see how to use UQLM for various hallucination detection methods:
- [Black-Box Uncertainty Quantification](https://github.com/cvs-health/uqlm/blob/develop/examples/black_box_demo.ipynb): A notebook demonstrating hallucination detection with black-box (consistency) scorers.
- [White-Box Uncertainty Quantification](https://github.com/cvs-health/uqlm/blob/develop/examples/white_box_demo.ipynb): A notebook demonstrating hallucination detection with white-box (token probability-based) scorers.
- [LLM-as-a-Judge](https://github.com/cvs-health/uqlm/blob/develop/examples/judges_demo.ipynb): A notebook demonstrating hallucination detection with LLM-as-a-Judge.
- [Tunable UQ Ensemble](https://github.com/cvs-health/uqlm/blob/develop/examples/ensemble_tuning_demo.ipynb): A notebook demonstrating hallucination detection with a tunable ensemble of UQ scorers ([Bouchard & Chauhan, 2023](https://arxiv.org/abs/2504.19254)).
- [Off-the-Shelf UQ Ensemble](https://github.com/cvs-health/uqlm/blob/develop/examples/ensemble_off_the_shelf_demo.ipynb): A notebook demonstrating hallucination detection using BS Detector ([Chen & Mueller, 2023](https://arxiv.org/abs/2308.16175)) off-the-shelf ensemble.
## Associated Research
A technical description of the `uqlm` scorers and extensive experiment results are contained in this **[this paper](https://arxiv.org/abs/2504.19254)**. If you use our framework or toolkit, we would appreciate citations to the following paper:
```bibtex
@misc{bouchard2025uncertaintyquantificationlanguagemodels,
title={Uncertainty Quantification for Language Models: A Suite of Black-Box, White-Box, LLM Judge, and Ensemble Scorers},
author={Dylan Bouchard and Mohit Singh Chauhan},
year={2025},
eprint={2504.19254},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2504.19254},
}
```