AI Paper Atlas

Research map / Large language models / Agents, code and math

BuildArena: A Physics-Aligned Interactive Benchmark of LLMs for Engineering Construction

Tian Xia, Tianrun Gao, Wenhao Deng, Long Wei, Xiaowei Qian, Chenglei Yu, Tailin Wu

ICML 2026 · Large language models › Agents, code and math · approach: Reasoning benchmarks and platforms · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add one strong example (one-shot) to the prompt (ablation)GPT-5 · Support Lv.1success rate (%): 85.9 -> 65.6 (↓20.3)Hurt
add one strong example (one-shot) to the prompt (ablation)GPT-4o · Support Lv.1success rate (%): 40.6 -> 25.0 (↓15.6)Hurt
add one strong example (one-shot) to the prompt (ablation)GPT-5 · Support Lv.1indicator (max load): 324.9 -> 294.0 (↓30.9)Hurt
add one strong example (one-shot) to the prompt (ablation)GPT-4o · Support Lv.1indicator (max load): 181.2 -> 160.6 (↓20.6)Hurt
add simulator feedback closed-loop retries (up to five rounds) (ablation)Fixed workflow · Support Lv.1success rate: to 100% success within five roundsImproved
add an analyst agent that reviews simulator trajectories and failure causes (ablation)Fixed workflow, three tested models · Lift Lv.2success rate: 0% -> non-zeroImproved
increase task difficulty from Lv.1 to Lv.3 in SupportGPT-5 · Supportsuccess rate (%): 85.9 -> 10.9Hurt
increase task difficulty from Lv.1 to Lv.3 in LiftGPT-5 · Liftsuccess rate (%): 95.3 -> 17.2Hurt
use more output tokens per construction runLLM construction runs · Transport, Support, Liftnormalized performanceNo clear effect
prompt explicitly for structural complexity and visual elaborateness (ablation)LLMs with the same module set · Construction complexitystructure intricacyImproved
scale up from GPT-4o-class models to GPT-5 on Lift Lv.2/3GPT-5 and Grok-4 vs others · Liftsuccess rateImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy