CodeSync: Multi-Agent System for Programming Assistance and Deterministic Thinking
Bibliographic record
Abstract
Large language models (LLMs) have transformed programming and software development, but their non-deterministic nature poses challenges to reproducible program synthesis. This paper proposes a multi-agent system, CodeSync, designed to align LLMs with human-like reasoning by employing a facilitator agent to coordinate tasks, alongside sub-agents responsible for thinking, planning, execution, validation, and evaluation. Utilizing strict decoding and robust reasoning validation, CodeSync emulates human cognitive processes-receiving a cue, devising a plan, breaking it into steps, adapting based on outcomes, and iterating toward the goal. This collaborative architecture not only outperforms existing multi-agent systems but also excels with open-source models, as demonstrated by CodeSync's impressive scores of 87.4% on the HumanEval benchmark and 91.2% on the MBPP (Mostly Basic Python Problems) benchmark with Llama-4-maverick, thereby eliminating the need for a strong reasoning model within the architecture.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.008 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.004 | 0.004 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.009 | 0.003 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".