Bye Bye Perspective API: Lessons for Measurement Infrastructure in NLP, CSS and LLM Evaluation
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866918471757987840 |
|---|---|
| author | Hartmann, David Tonneau, Manuel Kraft, Angelie Seiling, LK Staufer, Dimitri Delobelle, Pieter Fillies, Jan Luther, Anna Ricarda Batzner, Jan Lisker, Mareike |
| author_facet | Hartmann, David Tonneau, Manuel Kraft, Angelie Seiling, LK Staufer, Dimitri Delobelle, Pieter Fillies, Jan Luther, Anna Ricarda Batzner, Jan Lisker, Mareike |
| contents | The closure of Perspective API at the end of 2026 discards what has functioned as the de facto standard for automated toxicity measurement in NLP, CSS, and LLM evaluation research. We document the structural dependence that the communities built on this single proprietary tool and discuss how this dependence caused epistemic problems that have affected - and will likely continue to affect - collective research efforts. Perspective's model was periodically updated without versioning or disclosure, its annotation structure reflected a single corporate operationalisation of a contested concept, and its scores were used simultaneously as an evaluation target and an evaluation standard. Its closure leaves behind non-updatable benchmarks, irreproducible results, and ultimately a field at risk of perpetuating these issues by turning to closed-source LLMs. We use Perspective's announced termination as an opportunity to call for an independent, valid, adaptable, and reproducible toxicity and hate speech measurement infrastructure, with the technical and governance requirements outlined in this paper. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_25580 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Bye Bye Perspective API: Lessons for Measurement Infrastructure in NLP, CSS and LLM Evaluation Hartmann, David Tonneau, Manuel Kraft, Angelie Seiling, LK Staufer, Dimitri Delobelle, Pieter Fillies, Jan Luther, Anna Ricarda Batzner, Jan Lisker, Mareike Computation and Language I.2.7; K.4.1 The closure of Perspective API at the end of 2026 discards what has functioned as the de facto standard for automated toxicity measurement in NLP, CSS, and LLM evaluation research. We document the structural dependence that the communities built on this single proprietary tool and discuss how this dependence caused epistemic problems that have affected - and will likely continue to affect - collective research efforts. Perspective's model was periodically updated without versioning or disclosure, its annotation structure reflected a single corporate operationalisation of a contested concept, and its scores were used simultaneously as an evaluation target and an evaluation standard. Its closure leaves behind non-updatable benchmarks, irreproducible results, and ultimately a field at risk of perpetuating these issues by turning to closed-source LLMs. We use Perspective's announced termination as an opportunity to call for an independent, valid, adaptable, and reproducible toxicity and hate speech measurement infrastructure, with the technical and governance requirements outlined in this paper. |
| title | Bye Bye Perspective API: Lessons for Measurement Infrastructure in NLP, CSS and LLM Evaluation |
| topic | Computation and Language I.2.7; K.4.1 |
| url | https://arxiv.org/abs/2604.25580 |