Should I disclose my dataset? Caveats between reproducibility and individual data rights
Raysa M. Benatti | Camila M. L. Villarroel | Sandra Avila | Esther L. Colombini | Fabiana Severi
Proceedings of the Natural Legal Language Processing Workshop 2022
Natural language processing techniques have helped domain experts solve legal problems. Digital availability of court documents increases possibilities for researchers, who can access them as a source for building datasets — whose disclosure is aligned with good reproducibility practices in computational research. Large and digitized court systems, such as the Brazilian one, are prone to be explored in that sense. However, personal data protection laws impose restrictions on data exposure and state principles about which researchers should be mindful. Special caution must be taken in cases with human rights violations, such as gender discrimination, over which we elaborate as an example of interest. We present legal and ethical considerations on the issue, as well as guidelines for researchers dealing with this kind of data and deciding whether to disclose it.
CIDEr-R: Robust Consensus-based Image Description Evaluation
Gabriel Oliveira dos Santos | Esther Luna Colombini | Sandra Avila
Proceedings of the Seventh Workshop on Noisy User-generated Text (W-NUT 2021)
This paper shows that CIDEr-D, a traditional evaluation metric for image description, does not work properly on datasets where the number of words in the sentence is significantly greater than those in the MS COCO Captions dataset. We also show that CIDEr-D has performance hampered by the lack of multiple reference sentences and high variance of sentence length. To bypass this problem, we introduce CIDEr-R, which improves CIDEr-D, making it more flexible in dealing with datasets with high sentence length variance. We demonstrate that CIDEr-R is more accurate and closer to human judgment than CIDEr-D; CIDEr-R is more robust regarding the number of available references. Our results reveal that using Self-Critical Sequence Training to optimize CIDEr-R generates descriptive captions. In contrast, when CIDEr-D is optimized, the generated captions’ length tends to be similar to the reference length. However, the models also repeat several times the same word to increase the sentence length.