Description du projet
Approches neuro-symboliques pour les systèmes de NLG
La programmation IA peut être utilisée pour produire du texte à partir de données. Baptisée «génération du langage naturel» (NLG pour natural language generation), cette technique permet de transformer des données complexes en un langage à sonorité naturelle – comme s’il avait été écrit par un humain. Le projet NG-NLG, financé par l’UE, se penchera sur les approches neuronales de la NLG, qui demeurent actuellement confinées à un usage expérimental. En effet, malgré les résultats très naturels des systèmes neuronaux récents, le comportement des systèmes neuronaux NLG n’est ni transparent ni fiable. Le projet développera des approches innovantes combinant approches neuronales et représentations sémantiques symboliques explicites, offrant ainsi un meilleur contrôle des résultats et des inférences logiques explicites sur les données. Le projet testera ses approches sur la génération de données en texte, le résumé et la génération de réponses de dialogue.
Objectif
This project aims to overcome the major hurdles that prevent current state-of-the-art models for natural language generation (NLG) from real-world deployment.
While deep learning and neural networks brought considerable progress in many areas of natural language processing, neural approaches to NLG remain confined to experimental use and production NLG systems are handcrafted. The reason for this is that despite the very natural and fluent outputs of recent neural systems, neural NLG still has major drawbacks: (1) the behavior of the systems is not transparent and hard to control (the internal representation is implicit), which leads to incorrect or even harmful outputs, (2) the models require a lot of training data and processing power do not generalize well, and are mostly English-only. On the other hand, handcrafted models are safe, transparent and fast, but produce less fluent outputs and are expensive to adapt to new languages and domains (topics). As a result, usefulness of NLG models in general is limited. In addition, current methods for automatic evaluation of NLG outputs are unreliable, hampering system development.
The main aims of this project, directly addressing the above drawbacks, are:
1) Develop new approaches for NLG that combine neural approaches with explicit symbolic semantic representations, thus allowing greater control over the outputs and explicit logical inferences over the data.
2) Introduce approaches to model compression and adaptation to make models easily portable across domains and languages.
3) Develop reliable neural-symbolic approaches for evaluation of NLG systems.
We will test our approaches on multiple NLG applications—data-to-text generation (e.g. weather or sports reports), summarization, and dialogue response generation. For example, our approach will make it possible to deploy a new data reporting system for a given domain based on a few dozen example input-output pairs, compared to thousands needed by current methods.
Champ scientifique
Programme(s)
- HORIZON.1.1 - European Research Council (ERC) Main Programme
Thème(s)
Régime de financement
HORIZON-ERC - HORIZON ERC GrantsInstitution d’accueil
116 36 Praha 1
Tchéquie