Solvita: Enhancing Large Language Models for Competitive Programming via Agentic Evolution

TL;DR

Solvita introduces an agentic evolution framework for LLMs that enables continuous learning and improves competitive programming performance without weight updates.

Contribution

It presents a novel multi-agent system with trainable knowledge networks that adapt through reinforcement signals, enhancing reasoning over time.

Findings

Achieves state-of-the-art results on multiple competitive programming benchmarks.

Nearly doubles the accuracy of single-pass code generation baselines.

Outperforms existing multi-agent frameworks in competitive programming tasks.

Abstract

Large language models (LLMs) still struggle with the rigorous reasoning demands of hard competitive programming. While recent multi-agent frameworks attempt to bridge this reliability gap, they remain fundamentally stateless: they rely on static retrieval and discard the valuable problem-solving and debugging experience gained from previous tasks. To address this, we present Solvita, an agentic evolution framework that enables continuous learning without requiring weight updates to the underlying LLM. Solvita reorganizes problem-solving into a closed-loop system of strategy selection, program synthesis, certified supervision, and targeted hacking, executed by four specialized agents: Planner, Solver, Oracle, and Hacker. Crucially, each agent is paired with a trainable, graph-structured knowledge network. As the system operates, outcome signals, such as pass/fail verdicts, test certification quality, and adversarial vulnerabilities discovered by the Hacker, are recast as reinforcement learning updates to these network weights. This allows the agents to dynamically route future queries based on past successes and failures, effectively accumulating transferable reasoning experience over time. Evaluated across CodeContests, APPS, AetherCode, and live Codeforces rounds, Solvita establishes a new state-of-the-art among code-generation agents, outperforming existing multi-agent pipelines and nearly doubling the accuracy of single-pass baselines.