Hybrid-Gym: Training Coding Agents to Generalize Across Tasks
When assessing the quality of coding agents, predominant benchmarks focus on solving single issues on GitHub, such as SWE-Bench. In contrast, in real use, these agents solve more various and complex tasks that involve other skills such as exploring codebases, testing software, and designing architecture. In this paper, we first characterize some transferable skills that are shared across diverse tasks by decomposing trajectories into fine-grained components, and derive a set of principles for designing auxiliary training tasks to teach language models these skills. Guided by these principles, we propose a training environment, Hybrid-Gym, consisting of a set of scalable synthetic tasks, such as function localization and dependency search. Experiments show that agents trained on our synthetic tasks effectively generalize to diverse real-world tasks that are not present in training, improving a base model by 25.4% absolute gain on SWE-Bench Verified, 7.9% on SWT-Bench Verified, and 5.1% on Commit-0 Lite. Hybrid-Gym also complements datasets built for the downstream tasks (e.g., improving SWE-Play by 4.9% on SWT-Bench Verified). Code available at: https://github.com/yiqingxyq/Hybrid-Gym.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Coding Agents for Generalized Task and Motion Planning Problems
Task and motion planning (TAMP) problems remain difficult even with full observability and object-centric states because discrete decisions are tightly coupled to geometric, kinematic, and dynamic constraints. Generalize…
Program SynthesisMotion PlanningTEA: Trajectory Encoding Augmentation for Robust and Transferable Policies in Offline Reinforcement Learning
In this paper, we investigate offline reinforcement learning (RL) with the goal of training a single robust policy that generalizes effectively across environments with unseen dynamics. We propose a novel approach, Traje…
Reinforcement Learning (RL)A Hybrid Neuro-Symbolic approach for Text-Based Games using Inductive Logic Programming
Text-based games (TBGs) have emerged as an important test-bed, requiring reinforcement learning (RL) agents to combine natural language understanding with reasoning. A key challenge for agents solving this task is to gen…
Inductive logic programmingNatural Language UnderstandingReinforcement Learning (RL)text-based gamesHybrid Neural Coded Modulation: Design and Training Methods
We propose a hybrid coded modulation scheme which composes of inner and outer codes. The outer-code can be any standard binary linear code with efficient soft decoding capability (e.g. low-density parity-check (LDPC) cod…
CoAct-1: Computer-using Multi-Agent System with Coding Actions
Autonomous agents that operate computers via Graphical User Interfaces (GUIs) often struggle with efficiency and reliability on complex, long-horizon tasks. While augmenting these agents with planners can improve task de…