arxiv arXiv cs.CL · 1h ago · research

VisCAD releases foundation model suite with multimodal industrial CAD intelligence

from English

Researchers present VisCAD, a foundation model suite designed to provide broad generalization and strong capability for realistic industrial product design. The suite addresses the challenges of part-level generation from diverse inputs like renders and text, as well as assembly-level generation involving interacting parts.

VisCAD-M1 is a 27B model trained for part-level design generation that achieves an average score of 0.5540 on PubCADBench and RealCADBench.
This performance surpasses the strongest frontier model, which scored 0.5496.
Reusing VisCAD-M1 as a test-time verifier raises the score to 0.5797, representing an approximately 5 percent relative improvement over the previous state of the art.
The suite includes a domain-specific harness leveraging frontier models for complex assembly generation, demonstrating advantages over general-purpose harnesses.

VisCAD aims to overcome the limitations of existing specialized CAD models that generalize poorly and general-purpose models that perform inconsistently across CAD domains.

Importance 2/3 New harness with differentiators arXiv cs.CL Alibaba (Qwen) Multimodal Research paper

Related articles

arxiv arXiv cs.AI · 52d ago · 4 views

Test-time scaling improves small VLMs by fixing prompt parseability and increasing token budget

The study examines whether test-time scaling (TTS) transfers to small open vision-language models using the EXAMS-V multilingual visual multiple-choice benchmark. It compares self-consistency, describe-then-reason with PRM-guided beam search, and post-hoc selectors across Qwen2.5-VL-7B-Instruct and Qwen3.5-4B.

arxiv arXiv cs.CL · 53d ago · 1 view

Test-time scaling improves multilingual visual MCQ for small VLMs

The study examines whether test-time scaling (TTS) transfers to small open vision-language models using the EXAMS-V benchmark, comparing methods across Qwen2.5-VL-7B-Instruct and Qwen3.5-4B.

arxiv arXiv cs.CL · 73d ago · 4 views

BioMatrix: First Natively Multimodal Biological Foundation Model

BioMatrix integrates sequences, structures, and language for molecules and proteins in a single decoder-only architecture. It achieves state-of-the-art or competitive performance on 77 out of 80 downstream tasks, demonstrating effective multimodal generalist capabilities without external components.

arxiv arXiv cs.LG · 2d ago · 5 views

CANOPY enables Qwen3-14B to top AppWorld using outcome-only RL

The paper introduces CANOPY (Coverage-ANchored On-PolicY RL), a protocol that addresses signal starvation and policy drift in long-horizon reinforcement learning for small open models. By scaling same-task exploration and keeping updates KL-anchored, the method allows agents to learn effectively from end-of-task verification alone.

media MarkTechPost · 8d ago · 28 views

Alibaba's Qwen Team releases Qwen3.8-Flash-Next, a 125B MoE previewing Qwen4

Alibaba’s Qwen team has released Qwen3.8-Flash-Next, an open-weight multimodal Mixture-of-Experts model designed to preview the architecture for the upcoming Qwen4 series. The checkpoint pairs a 125B backbone with a 51B N-gram embedding table and a 4B multi-token prediction module, activating only 6B parameters per token.