AI Paper Atlas

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

From Assistant to Independent Developer — Are GPTs Ready for Software Development?

Dezhi Ran, Yuan Cao, Mengzhou Wu, Simin Chen, Yuzhe Guo, Jun Ren, Zihe Song, Hao Yu, Jialei Wei, Linyi Li, Wei Yang, Baishakhi Ray and 1 more

ICLR 2026 · Large language models › Agents, code and math · approach: Agent instruction-following benchmarks

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 compilation error feedback (repair round) (ablation)Claude-4-Sonnet · App-Forge compile ratecompile rate (%): 40.59% -> 77.23%Improved
add compilation error feedback (repair round) (ablation)Qwen3-Coder · App-Forge compile ratecompile rate (%): 27.72% -> 85.15%Improved
add compilation error feedback (repair round) (ablation)GPT-5-High · App-Forge functional successfunctional success rate (%): 14.85% -> 18.81%Improved
add compilation error feedback (repair round) (ablation)Claude-4-Sonnet · App-Forge functional successfunctional success rate (%): 0.99% -> 3.96%Improved
add compilation error feedback (repair round) (ablation)GPT-4.1 · App-Forge functional successfunctional success rate (%): 0.99% -> 0.99%No clear effect
add compilation error feedback (repair round) (ablation)GPT-4.1 · App-Forge lines of codeLOC: 367.43 -> 58.41Hurt
iterate compilation-feedback repair for 2-3 rounds (ablation)Qwen3-Coder · App-Forge functional successfunctional success rate (%): saturates after 2-3 iterations, peaking around 23%No clear effect
increase GPT-5 reasoning level from low to high (ablation)GPT-5 · App-Forge functional successfunctional success rate (%): 2.97% -> 14.85%Improved
wrap model in mini-SWE-agent instead of plain generation (ablation)Claude-4-Opus · App-Forge functional successfunctional success rate (%): 11.88% -> 11.88%No clear effect
use mini-SWE-agent instead of plain generation (ablation)Qwen3-Coder · App-Forge functional successfunctional success rate (%): 1.98% -> 6.93%Improved
add compilation feedback (ablation)GPT-4.1 · App-Forge runtime failed to startfailed to start rate: 2.0 -> 66.0Hurt

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