Yoav Sterman

Yoav Sterman

Auto-generated profile, based on publicly available information

Yoav Sterman


Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
8 February 2023 in Additive Manufacturing
Verified
Journal Article
Closed access
AHAnat Halbrecht
Anat Halbrecht
MKMaayan Kinsbursky
Maayan Kinsbursky
RPRoi Poranne
Roi Poranne
YSYoav Sterman
Yoav Sterman

2 February 2022 in Computer-Aided Design
Verified
Journal Article
Closed access
YSYoav Sterman
Yoav Sterman
EAEli Almog
Eli Almog


23 June 2025 in International Journal of Architectural Computing
Verified
Journal Article
Open access
BHBat El Hizmi
Bat El Hizmi
YSYoav Sterman
Yoav Sterman
Guy AusternGuy Austern
Guy Austern
Recent advancements in Machine Learning (ML) have significantly enhanced the capability to generate 3D objects from textual descriptions, offering significant potential for design and fabrication workflows. However, these models typically fail to meet practical requirements like printability or manufacturability. In addition, current approaches often cannot accurately control the dimensions and the interrelations of elements within the generated 3D models. This presents a major challenge in utilizing ML-generated designs in real-world applications. To address this gap, we introduce a novel method for translating natural language descriptions into parametric 3D objects using Large Language Models (LLMs). Our approach employs multiple agents, each agent is LLM pre-trained for a specific task. The first agent deconstructs textual prompts into design elements and describes their geometry and spatial relations. The second agent translates the description into code using the Rhino.Geometry coding library in the Rhino3D-Grasshopper modeling environment. A final agent reassembles the models and adds parametric control interfaces, enabling customizable outputs. In this paper, we describe the method’s architecture and the training methodologies, including in-context learning and fine-tuning. The results demonstrate that the suggested method successfully generates code for variations of familiar objects. However, it still struggles with more complex designs that deviate significantly from the training data. We outline future directions for improvement, including expanding the training dataset and exploring advanced LLM models. This work is a step towards making 3D modeling accessible to a broader audience, using everyday language to simplify design processes.

19 April 2023
Verified
Conference Paper
Open access
SAShahar Asor
Shahar Asor
YSYoav Sterman
Yoav Sterman
Punchcard knitting machines allow automatic needle selection based on a punched pattern. The machine reads the card and translates a punched hole to specific knitting actions. Punchcards are usually purchased ready to use or manually punched by the user. Current tools allow the digital fabrication of punchcards, however those tools focus on two-dimensional graphic patterns. This work introduces a parametric workflow for designing and fabricating custom-made punchcards that enable 3D knitting and shaping. We demonstrate our workflow using two case studies that implement short-row knitting to obtain 3D surfaces and non-rectangular shapes. In the first, we generate parametric spike 3D textures, and in the second, we demonstrate long curved shapes and 3D cones. A DXF file is saved and used for cutting the punchcard using a digital desktop cutting machine. Our workflow broadens the design opportunities when using punchcard-driven domestic knitting machines.