Analyzing constrained LLM through PDFA-learning
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909224526675968 |
|---|---|
| author | Carrasco, Matías Mayr, Franz Yovine, Sergio Kidd, Johny Iturbide, Martín da Silva, Juan Pedro Garat, Alejo |
| author_facet | Carrasco, Matías Mayr, Franz Yovine, Sergio Kidd, Johny Iturbide, Martín da Silva, Juan Pedro Garat, Alejo |
| contents | We define a congruence that copes with null next-symbol probabilities that arise when the output of a language model is constrained by some means during text generation. We develop an algorithm for efficiently learning the quotient with respect to this congruence and evaluate it on case studies for analyzing statistical properties of LLM. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_08269 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Analyzing constrained LLM through PDFA-learning Carrasco, Matías Mayr, Franz Yovine, Sergio Kidd, Johny Iturbide, Martín da Silva, Juan Pedro Garat, Alejo Formal Languages and Automata Theory Artificial Intelligence Machine Learning We define a congruence that copes with null next-symbol probabilities that arise when the output of a language model is constrained by some means during text generation. We develop an algorithm for efficiently learning the quotient with respect to this congruence and evaluate it on case studies for analyzing statistical properties of LLM. |
| title | Analyzing constrained LLM through PDFA-learning |
| topic | Formal Languages and Automata Theory Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2406.08269 |