MEMRES: A Memory-Augmented Resolver with Confidence Cascade for Agentic Python Dependency Resolution

TL;DR

MEMRES is an advanced Python dependency resolver that combines memory, error knowledge, semantic analysis, heuristics, and LLMs to achieve high success rates in resolving code snippets.

Contribution

It introduces a multi-level confidence cascade and a self-evolving memory system that significantly improve dependency resolution success compared to previous methods.

Findings

Resolves 86.6% of code snippets on HG2.9K dataset.

Outperforms PLLM's 54.7% success rate by a wide margin.

Utilizes a knowledge base of 200+ import-to-package mappings.

Abstract

We present MEMRES, an agentic system for Python dependency resolution that introduces a multi-level confidence cascade where the LLM serves as the last resort. Our system combines: (1) a Self-Evolving Memory that accumulates reusable resolution patterns via tips and shortcuts; (2) an Error Pattern Knowledge Base with 200+ curated import-to-package mappings; (3) a Semantic Import Analyzer; and (4) a Python 2 heuristic detector resolving the largest failure category. On HG2.9K using Gemma-2 9B (10 GB VRAM). MEMRES resolves 2503 of 2890 (86.6%, 10-run average) snippets, combining intra-session memory with our confidence cascade for the remainder. This already exceeds PLLM's 54.7% overall success rate by a wide margin.