Skip to main content
Przejdź do strony domowej Komisji Europejskiej (odnośnik otworzy się w nowym oknie)
polski pl
CORDIS - Wyniki badań wspieranych przez UE
CORDIS

From scaling to eFficiency LAws for viSual syntHesis

Cel

Current AI foundation models are trained on the entire Internet, consume enormous GPU resources, and follow the “scale is god” paradigm: the more data and compute, the better the results. Yet despite their scale, such models rarely produce the desired output on the first attempt, forcing users into repeated trial-and-error interactions. This inefficiency is most evident in conditional visual generation, where even lengthy, fine-grained text prompts rarely yield the desired image or video.

FLASH proposes a paradigm shift: from scale to efficiency.
We ask a fundamental question: Can we build models that reach high-quality, controllable visual outputs through the most efficient path, using the least data, compute, and interaction?

We propose three efficiency resources with concrete methods for each:
First, data efficiency: identifying how data quality, diversity, and type can replace raw volume.
Second, data–compute efficiency: designing models whose capacity and training follow the data, using flow-matching with adaptive extra conditions.
Third, interaction efficiency: leveraging multimodal user-model interaction by modelling abstract intermediates, such as sketch or layout; a user intent state; and targeted questions to reduce user–model turns.

Ultimately, our goal is to uncover and learn efficiency laws that govern conditional visual generation and use them to build the next generation of multimodal foundation models, efficient by design.

We will show impact on three applications, where efficiency is a game changer: (1) text-to-image generation, (2) multimodal generation and (3) robot manipulation, operating under strict constraints (memory, energy, latency).

By replacing the race to scale with principled efficiency, FLASH will democratise conditional generation, promote sustainability, and enable deployment on constrained environments such as edge devices or robotics. We hope that FLASH will help create new models across diverse domains, accessible to all.

Dziedzina nauki (EuroSciVoc)

Projekt nie został jeszcze sklasyfikowany według klasyfikacji EuroSciVoc.
Wskaż dziedziny nauki, które twoim zdaniem są najbardziej istotne z punktu widzenia tego projektu i pomóż nam usprawnić naszą usługę klasyfikacji.

Słowa kluczowe

Program(-y)

Temat(-y)

System finansowania

HORIZON-ERC - HORIZON ERC Grants

Wyświetl wszystkie projekty finansowane w ramach tego programu finansowania

Zaproszenie do składania wniosków

(odnośnik otworzy się w nowym oknie) ERC-2026-STG

Wyświetl wszystkie projekty finansowane w ramach tego zaproszenia

Instytucja przyjmująca

ECOLE POLYTECHNIQUE
Wkład UE netto
€ 1 499 990,00
Koszt całkowity
€ 1 499 990,00

Beneficjenci (1)