POLAR:A Per-User Association Test in Embedding Space
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911520975224832 |
|---|---|
| author | Bento, Pedro Buzelin, Arthur Chagas, Arthur Aquino, Yan Estanislau, Victoria Malaquias, Samira Dutenhefner, Pedro Robles Pappa, Gisele L. Almeida, Virgilio MeiraJr, Wagner |
| author_facet | Bento, Pedro Buzelin, Arthur Chagas, Arthur Aquino, Yan Estanislau, Victoria Malaquias, Samira Dutenhefner, Pedro Robles Pappa, Gisele L. Almeida, Virgilio MeiraJr, Wagner |
| contents | Most intrinsic association probes operate at the word, sentence, or corpus level, obscuring author-level variation. We present POLAR (Per-user On-axis Lexical Association Re-port), a per-user lexical association test that runs in the embedding space of a lightly adapted masked language model. Authors are represented by private deterministic to-kens; POLAR projects these vectors onto curated lexicalaxes and reports standardized effects with permutation p-values and Benjamini--Hochberg control. On a balanced bot--human Twitter benchmark, POLAR cleanly separates LLM-driven bots from organic accounts; on an extremist forum,it quantifies strong alignment with slur lexicons and reveals rightward drift over time. The method is modular to new attribute sets and provides concise, per-author diagnostics for computational social science. All code is publicly avail-able at https://github.com/pedroaugtb/POLAR-A-Per-User-Association-Test-in-Embedding-Space. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_15950 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | POLAR:A Per-User Association Test in Embedding Space Bento, Pedro Buzelin, Arthur Chagas, Arthur Aquino, Yan Estanislau, Victoria Malaquias, Samira Dutenhefner, Pedro Robles Pappa, Gisele L. Almeida, Virgilio MeiraJr, Wagner Computation and Language Computers and Society Social and Information Networks Most intrinsic association probes operate at the word, sentence, or corpus level, obscuring author-level variation. We present POLAR (Per-user On-axis Lexical Association Re-port), a per-user lexical association test that runs in the embedding space of a lightly adapted masked language model. Authors are represented by private deterministic to-kens; POLAR projects these vectors onto curated lexicalaxes and reports standardized effects with permutation p-values and Benjamini--Hochberg control. On a balanced bot--human Twitter benchmark, POLAR cleanly separates LLM-driven bots from organic accounts; on an extremist forum,it quantifies strong alignment with slur lexicons and reveals rightward drift over time. The method is modular to new attribute sets and provides concise, per-author diagnostics for computational social science. All code is publicly avail-able at https://github.com/pedroaugtb/POLAR-A-Per-User-Association-Test-in-Embedding-Space. |
| title | POLAR:A Per-User Association Test in Embedding Space |
| topic | Computation and Language Computers and Society Social and Information Networks |
| url | https://arxiv.org/abs/2603.15950 |