You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
UQLM is a Python library for Large Language Model (LLM) hallucination detection using state-of-the-art uncertainty quantification techniques.
19
49
@@ -310,21 +340,22 @@ Each notebook includes detailed explanations and code samples that you can adapt
310
340
311
341
312
342
## Citation
313
-
A technical description of the `uqlm` scorers and extensive experimental results are presented in **[this paper](https://arxiv.org/abs/2504.19254)**. If you use our framework or toolkit, please cite:
343
+
A technical description of the `uqlm` scorers and extensive experimental results are presented in **[this paper](https://openreview.net/pdf?id=WOFspd4lq5)**, published in **Transactions on Machine Learning Research (TMLR)**. If you use our framework or toolkit, please cite:
title={Uncertainty Quantification for Language Models: A Suite of Black-Box, White-Box, LLM Judge, and Ensemble Scorers},
318
-
author={Dylan Bouchard and Mohit Singh Chauhan},
319
-
year={2025},
320
-
eprint={2504.19254},
321
-
archivePrefix={arXiv},
322
-
primaryClass={cs.CL},
323
-
url={https://arxiv.org/abs/2504.19254},
346
+
@article{
347
+
bouchard2025uncertainty,
348
+
title={Uncertainty Quantification for Language Models: A Suite of Black-Box, White-Box, {LLM} Judge, and Ensemble Scorers},
349
+
author={Dylan Bouchard and Mohit Singh Chauhan},
350
+
journal={Transactions on Machine Learning Research},
351
+
issn={2835-8856},
352
+
year={2025},
353
+
url={https://openreview.net/forum?id=WOFspd4lq5},
354
+
note={}
324
355
}
325
356
```
326
357
327
-
The `uqlm` software package is described in this **[this paper](https://arxiv.org/abs/2507.06196)**. If you use the software, please cite:
358
+
The `uqlm` software package is described in this **[this paper](https://arxiv.org/abs/2507.06196)**, published in the **Journal of Machine Learning Research (JMLR)**. If you use the software, please cite:
328
359
329
360
```bibtex
330
361
@misc{bouchard2025uqlmpythonpackageuncertainty,
@@ -337,3 +368,16 @@ The `uqlm` software package is described in this **[this paper](https://arxiv.or
337
368
url={https://arxiv.org/abs/2507.06196},
338
369
}
339
370
```
371
+
372
+
The long-text methods and experiment results are described in **this paper**, available as a preprint on arXiv. To cite:
0 commit comments