
An LLM-based multi-module agent that turns a single natural-language description into a fully populated, interactive healthcare VR scene in Unity, with a frontier-LLM planner, lightweight VLM critics, and an LLM-free runtime interaction layer.
Oct 1, 2026

PDAGENT-BENCH is a comprehensive benchmark of 353 curated problems for evaluating LLM/VLM agents across the VLSI physical design stack, revealing a substantial gap between conceptual understanding and tool-grounded execution.
Jun 15, 2026