Search papers, labs, and topics across Lattice.
The paper introduces LiteCoder-Terminal-Gen, a zero-dependency pipeline for generating executable terminal environments from domain specifications, addressing limitations of existing scraped datasets. They create two datasets, LiteCoder-Terminal-SFT (expert trajectories) and LiteCoder-Terminal-RL (verifiable environments), and fine-tune Qwen models on them. Results show significant performance gains on Terminal Bench benchmarks after supervised fine-tuning and further improvements with Direct Multi-turn Preference Optimization (DMPO), demonstrating the effectiveness of synthetic environments for training language agents in complex command-line tasks.
Forget scraping – this work shows you can generate high-quality, executable terminal environments from scratch to train language agents that outperform models trained on scraped data.
Mastering terminal environments requires language agents capable of multi-step planning, feedback-grounded execution, and dynamic state adaptation. However, training such agents is currently bottlenecked by a reliance on scraped external repositories, which limits domain diversity, environment controllability, and the targeting of specific capability deficits. We introduce LiteCoder-Terminal-Gen, a zero-dependency synthesis pipeline that autonomously generates executable and verifiable terminal training environments directly from domain specifications. Using this framework, we construct two large-scale resources: LiteCoder-Terminal-SFT, comprising 11,255 expert trajectories across 10 domains, and LiteCoder-Terminal-RL, featuring 602 verifiable environments for trajectory-level preference optimization. Supervised fine-tuning of Qwen-family models on our SFT dataset yields agents that significantly outperform their base counterparts. Notably, our 32B variant achieves 29.06%, 18.54%, and 34.00% pass@1 on Terminal Bench 1.0, 2.0, and Pro, respectively. Furthermore, applying Direct Multi-turn Preference Optimization (DMPO) on our RL environments yields additional performance gains. These results systematically demonstrate that fully synthetic, executable environments offer a scalable and verifiable supervision signal for mastering complex, real-world command-line workflows.
