Large Concept Models: Language Modeling in a Sentence Representation Space
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866929632261963776 |
|---|---|
| author | LCM team Barrault, Loïc Duquenne, Paul-Ambroise Elbayad, Maha Kozhevnikov, Artyom Alastruey, Belen Andrews, Pierre Coria, Mariano Couairon, Guillaume Costa-jussà, Marta R. Dale, David Elsahar, Hady Heffernan, Kevin Janeiro, João Maria Tran, Tuan Ropers, Christophe Sánchez, Eduardo Roman, Robin San Mourachko, Alexandre Saleem, Safiyyah Schwenk, Holger |
| author_facet | LCM team Barrault, Loïc Duquenne, Paul-Ambroise Elbayad, Maha Kozhevnikov, Artyom Alastruey, Belen Andrews, Pierre Coria, Mariano Couairon, Guillaume Costa-jussà, Marta R. Dale, David Elsahar, Hady Heffernan, Kevin Janeiro, João Maria Tran, Tuan Ropers, Christophe Sánchez, Eduardo Roman, Robin San Mourachko, Alexandre Saleem, Safiyyah Schwenk, Holger |
| contents | LLMs have revolutionized the field of artificial intelligence and have emerged as the de-facto tool for many tasks. The current established technology of LLMs is to process input and generate output at the token level. This is in sharp contrast to humans who operate at multiple levels of abstraction, well beyond single words, to analyze information and to generate creative content. In this paper, we present an attempt at an architecture which operates on an explicit higher-level semantic representation, which we name a concept. Concepts are language- and modality-agnostic and represent a higher level idea or action in a flow. Hence, we build a "Large Concept Model". In this study, as proof of feasibility, we assume that a concept corresponds to a sentence, and use an existing sentence embedding space, SONAR, which supports up to 200 languages in both text and speech modalities.
The Large Concept Model is trained to perform autoregressive sentence prediction in an embedding space. We explore multiple approaches, namely MSE regression, variants of diffusion-based generation, and models operating in a quantized SONAR space. These explorations are performed using 1.6B parameter models and training data in the order of 1.3T tokens. We then scale one architecture to a model size of 7B parameters and training data of about 2.7T tokens. We perform an experimental evaluation on several generative tasks, namely summarization and a new task of summary expansion. Finally, we show that our model exhibits impressive zero-shot generalization performance to many languages, outperforming existing LLMs of the same size. The training code of our models is freely available. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_08821 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Large Concept Models: Language Modeling in a Sentence Representation Space LCM team Barrault, Loïc Duquenne, Paul-Ambroise Elbayad, Maha Kozhevnikov, Artyom Alastruey, Belen Andrews, Pierre Coria, Mariano Couairon, Guillaume Costa-jussà, Marta R. Dale, David Elsahar, Hady Heffernan, Kevin Janeiro, João Maria Tran, Tuan Ropers, Christophe Sánchez, Eduardo Roman, Robin San Mourachko, Alexandre Saleem, Safiyyah Schwenk, Holger Computation and Language LLMs have revolutionized the field of artificial intelligence and have emerged as the de-facto tool for many tasks. The current established technology of LLMs is to process input and generate output at the token level. This is in sharp contrast to humans who operate at multiple levels of abstraction, well beyond single words, to analyze information and to generate creative content. In this paper, we present an attempt at an architecture which operates on an explicit higher-level semantic representation, which we name a concept. Concepts are language- and modality-agnostic and represent a higher level idea or action in a flow. Hence, we build a "Large Concept Model". In this study, as proof of feasibility, we assume that a concept corresponds to a sentence, and use an existing sentence embedding space, SONAR, which supports up to 200 languages in both text and speech modalities. The Large Concept Model is trained to perform autoregressive sentence prediction in an embedding space. We explore multiple approaches, namely MSE regression, variants of diffusion-based generation, and models operating in a quantized SONAR space. These explorations are performed using 1.6B parameter models and training data in the order of 1.3T tokens. We then scale one architecture to a model size of 7B parameters and training data of about 2.7T tokens. We perform an experimental evaluation on several generative tasks, namely summarization and a new task of summary expansion. Finally, we show that our model exhibits impressive zero-shot generalization performance to many languages, outperforming existing LLMs of the same size. The training code of our models is freely available. |
| title | Large Concept Models: Language Modeling in a Sentence Representation Space |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2412.08821 |