Search papers, labs, and topics across Lattice.
This paper introduces PairCoder, a novel approach to code-driven generation of structured artifacts through a two-agent pair programming paradigm, where a Driver agent writes code and a Navigator agent reviews it against verification evidence. By leveraging this interactive method, PairCoder significantly enhances the performance of large language models across 17 public benchmarks, achieving substantial improvements in artifact verifiability metrics. The results show that PairCoder can outperform traditional single-pass inference methods, particularly in scenarios where the toolchain provides strong verification signals, demonstrating the efficacy of collaborative code generation in AI applications.
PairCoder boosts artifact verifiability by up to 3.9 times compared to traditional single-pass inference, revealing the power of collaborative programming in AI-generated outputs.
Code is the medium through which large language models generate structured artifacts: charts, scientific figures, vector graphics, CAD models, 3D scenes, and hardware designs are all produced by writing programs. In this regime single pass inference is brittle, because the compiler, renderer, or simulator that decides whether the artifact exists is invisible to the model. We present PairCoder, which grounds review in the toolchain and realizes it as two agent pair programming: a Driver agent writes the program, a Navigator agent reviews it against verification evidence (diagnostics, execution results, and renderings of the current artifact beside the target), and the two switch roles when errors persist. Across 17 public benchmarks and seven models from three vendors, PairCoder improves essentially every benchmark whose artifact is verifiable, on full official metric suites rather than execution alone (for example, Blender scene executability 0.20 to 0.78; TikZ compile rate up 10 to 30 points on every model), at 2.9 to 9.2 times single model cost (about 7 times overall). The improvements concentrate where the toolchain provides an informative oracle and the baseline leaves headroom, and the method ties or mildly regresses where the oracle is weak; we frame pair programming as a reliable recipe for verified code driven generation.
