Yuling Shi

Repo0 redefines code generation by enabling agents to construct entire software architectures from scratch, achieving unprecedented functionality and reliability.

Proactively synthesizing project-specific issues can drastically enhance LLM agents' ability to resolve software problems effectively.

Temporal confidence allows for adaptive computation in parallel reasoning, leading to a 32% reduction in latency without sacrificing accuracy.

Current AI coding agents struggle with large-scale refactoring tasks, achieving only a 41.2% success rate on a newly curated benchmark designed to challenge their capabilities.

Pruning tool outputs directly within the agent leads to a remarkable 39% reduction in token usage without sacrificing performance.

ACQUIRE turns knowledge gaps into actionable insights, boosting software repair accuracy by over 4% while keeping costs low.

Benchmark scores for coding agents may mislead progress assessments, with only 39% of GSO tasks passing validity checks across machines.

Mirror-Fusion Attention reveals that lightweight geometric priors can significantly enhance self-supervised learning performance without the need for extensive model redesigns.

Dockerless achieves a 14.3 AUC point improvement in program verification without the overhead of Docker environments, revolutionizing efficiency in training coding agents.

FastContext cuts coding agent token usage by 60% while boosting resolution rates by 5.5% by decoupling code exploration from task-solving.

Integrating visual graphs with text interfaces allows LLM agents to reduce token consumption by 26% while enhancing their issue-resolution accuracy.