User Embedding Model for Personalized Language Prompting
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_ | 1866911753592373248 |
|---|---|
| author | Doddapaneni, Sumanth Sayana, Krishna Jash, Ambarish Sodhi, Sukhdeep Kuzmin, Dima |
| author_facet | Doddapaneni, Sumanth Sayana, Krishna Jash, Ambarish Sodhi, Sukhdeep Kuzmin, Dima |
| contents | Modeling long histories plays a pivotal role in enhancing recommendation systems, allowing to capture user's evolving preferences, resulting in more precise and personalized recommendations. In this study we tackle the challenges of modeling long user histories for preference understanding in natural language. Specifically, we introduce a new User Embedding Module (UEM) that efficiently processes user history in free-form text by compressing and representing them as embeddings, to use them as soft prompts to a LM. Our experiments demonstrate the superior capability of this approach in handling significantly longer histories compared to conventional text based prompting methods, yielding substantial improvements in predictive performance. The main contribution of this research is to demonstrate the ability to bias language models with user signals represented as embeddings. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_04858 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | User Embedding Model for Personalized Language Prompting Doddapaneni, Sumanth Sayana, Krishna Jash, Ambarish Sodhi, Sukhdeep Kuzmin, Dima Computation and Language Artificial Intelligence Information Retrieval Machine Learning Modeling long histories plays a pivotal role in enhancing recommendation systems, allowing to capture user's evolving preferences, resulting in more precise and personalized recommendations. In this study we tackle the challenges of modeling long user histories for preference understanding in natural language. Specifically, we introduce a new User Embedding Module (UEM) that efficiently processes user history in free-form text by compressing and representing them as embeddings, to use them as soft prompts to a LM. Our experiments demonstrate the superior capability of this approach in handling significantly longer histories compared to conventional text based prompting methods, yielding substantial improvements in predictive performance. The main contribution of this research is to demonstrate the ability to bias language models with user signals represented as embeddings. |
| title | User Embedding Model for Personalized Language Prompting |
| topic | Computation and Language Artificial Intelligence Information Retrieval Machine Learning |
| url | https://arxiv.org/abs/2401.04858 |