DARS: Dynamic Action Re-Sampling to Enhance Coding Agent Performance by Adaptive Tree Traversal
Large Language Models (LLMs) have revolutionized various domains, including natural language processing, data analysis, and software development, by enabling automation. In software engineering, LLM-powered coding agents have garnered significant attention due to their potential to automate complex development tasks, assist in debugging, and enhance productivity. However, existing approaches often struggle with sub-optimal decision-making, requiring either extensive manual intervention or inefficient compute scaling strategies. To improve coding agent performance, we present Dynamic Action Re-Sampling (DARS), a novel inference time compute scaling approach for coding agents, that is faster and more effective at recovering from sub-optimal decisions compared to baselines. While traditional agents either follow linear trajectories or rely on random sampling for scaling compute, our approach DARS works by branching out a trajectory at certain key decision points by taking an alternative action given the history of the trajectory and execution feedback of the previous attempt from that point. We evaluate our approach on SWE-Bench Lite benchmark, demonstrating that this scaling strategy achieves a pass@k score of 55% with Claude 3.5 Sonnet V2. Our framework achieves a pass@1 rate of 47%, outperforming state-of-the-art (SOTA) open-source frameworks.
Code (2)
Methods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
From Sampled Outcomes to Capability Distributions: Rethinking Supervision for LLM Routing
Existing LLM routing methods often construct supervision from a single sampled response for each query--model pair. Because LLM generation is stochastic, however, such an observation can be an unstable estimate of model …
DarSwin: Distortion Aware Radial Swin Transformer
Wide-angle lenses are commonly used in perception tasks requiring a large field of view. Unfortunately, these lenses produce significant distortions, making conventional models that ignore the distortion effects unable t…
Depth EstimationSmoothed Phase-Coded FMCW: Waveform Properties and Transceiver Architecture
Smoothed phase-coded frequency modulated continuous waveform (SPC-FMCW), which is aimed to improve the coexistence of multiple radars operating within the same frequency bandwidth, is studied, and the receiving strategy …
Computationally Discovered Potentiating Role of Glycans on NMDA Receptors
N-methyl-D-aspartate receptors (NMDARs) are glycoproteins in the brain central to learning and memory. The effects of glycosylation on the structure and dynamics of NMDARs are largely unknown. In this work, we use extens…
DarSwin-Unet: Distortion Aware Encoder-Decoder Architecture
Wide-angle fisheye images are becoming increasingly common for perception tasks in applications such as robotics, security, and mobility (e.g. drones, avionics). However, current models often either ignore the distortion…
DecoderDepth Estimation