AI Paper Atlas

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

MiniAppBench: Evaluating the Shift from Text to Interactive HTML Responses in LLM-Powered Assistants

Zuhao zhang, Chengyue Yu, Yuante Li, Chenyi Zhuang, Linjian Mo, Shuai Li

ICML 2026 · Large language models › Agents, code and math · approach: Software engineering agents · 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 source code as judge input (ablation)MiniAppEval · Agreement with human labels, 200-item subsetaccuracy (%): 70.66 -> 89.62Improved
add Playwright agent interaction (ablation)MiniAppEval · Agreement with human labels, 200-item subsetaccuracy (%): 66.48 -> 89.62Improved
add evaluation reference (ablation)MiniAppEval · Agreement with human labels, 200-item subsetaccuracy (%): 60.12 -> 89.62Improved
add evaluation reference (ablation)MiniAppEval · Agreement with human labels, 200-item subsetprecision (%): 89.47 -> 83.87Hurt
add evaluation reference (ablation)MiniAppEval · Agreement with human labels, 200-item subsetF1 (%): 61.08 -> 84.55Improved
add double-blind judging (judge sees output before query)MiniAppEval (Gemini-3-Pro-Preview) · Graphical queries (55), Gemini-3-Pro-Preview judgeaccuracy (%): 81.82 -> 85.45Improved
add double-blind judgingMiniAppEval (GPT-5.2) · Graphical queries (55), GPT-5.2 judgeaccuracy (%): 80.00 -> 83.63Improved
add double-blind judgingMiniAppEval (Claude-Opus-4.5) · Graphical queries (55), Claude-Opus-4.5 judgeaccuracy (%): 78.18 -> 83.63Improved
scale Qwen3 from 32B to 235B-A22BQwen3 · MiniAppBenchpass rate (%): 0.66 -> 2.88Improved
scale GLM from 4.5-Air to 4.7GLM · MiniAppBenchpass rate (%): 7.09 -> 18.31Improved
use closed-source instead of open-source modelsLLMs · MiniAppBenchavg pass rate (%): 6.19 -> 45.46Improved
raise task difficulty from Easy to Hardall models average · MiniAppBenchpass rate (%): 34.05 -> 4.34Hurt

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