zzzz7788990213-ops/EvoVLM

Evolving Multimodal Inference Programs: LLM-guided evolutionary search for accurate and efficient vision-language inference.

Open original source ↗

Older snapshot: check the original source for current information.

From the publisher

README & documentation

Source preview

Read the project’s overview, installation instructions and usage examples. The original README is the source of truth.

Evolve the inference program. Keep the model frozen. EvoVLM is a research project on LLM-guided, multi-objective evolutionary search for vision-language inference. It searches over prompts, image preprocessing, generation budgets, and answer extraction to improve the accuracy–latency trade-off on scientific chart understanding, without fine-tuning model weights. 中文:EvoVLM 在固定视觉语言模型权重的前提下,通过大模型引导的变异、实测评估和 Pareto 选择,迭代优化推理程序,在图表理解任务上探索准确率与推理效率的平衡。 Method · Results · Quick start · Technical report · Reproducibility The implementation follows the evolutionary-algorithm pattern of mutation, evaluation, and selection, with a constrained program representation: Qwen3-VL-2B-Instruct and Qwen3-VL-2B-Thinking. mutations. A trusted renderer turns the selected configuration into Python. correctness, latency, errors, and execution provenance. continue the next generation. The recorded search is deliberately small: two generations, with at most six children per model population. It is not unrestricted code generation, weight training, or a claim to implement a named algorithm such as NSGA-II in full. The implementation's distinguishing choices are constrained LLM-guided mutations, executable inference programs, and fail-closed evidence…

zzzz7788990213-ops/EvoVLM on GitHub A short preview, not the full document.

Read the full README ↗ · Preview checked 2026-11-13T20:20:43.998Z

Inside the original README — Document outline
  1. EvoVLM
  2. Evolving Multimodal Inference Programs
  3. Method
  4. Results
  5. Quick start
  6. Run inference
  7. Repository layout
  8. Reproducibility and scope
  9. Attribution and reuse

Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.

Links from the README

References supplied by the publisher, not independently verified endorsements. Check the destination before downloading files or entering credentials.

Documentation belongs to its respective authors. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.

What this repository does

Evolving Multimodal Inference Programs: LLM-guided evolutionary search for accurate and efficient vision-language inference.

Repository facts

Owner
zzzz7788990213-ops
Primary language
Python
Stars
508
Forks
5
Open issues + pull requests
0
License
Not reported — inspect the license file
Archived
No
Default branch
main
Created
2026-10-02T01:03:55.000Z
Last push
2026-10-02T01:04:42.000Z

Topics and intended use

No topics were included in the latest source metadata.

Review the README for scope, installation, examples and limitations. We do not run repository code or certify it.

Evaluate before installing

Review licensing and dependencies, inspect recent commits and unresolved issues, and test in an isolated environment before production use. Stars and forks alone cannot answer these questions.

README and project files

Issues and maintenance discussion

Releases and changelog

Source and freshness

Source: GitHub. Metadata observed 2026-10-09T12:06:54.953Z. Daily imports are snapshots, not real-time monitoring.

Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.

Open original source

Related repositories

public-apis/public-apis

EbookFoundation/free-programming-books

donnemartin/system-design-primer

vinta/awesome-python

practical-tutorials/project-based-learning

NousResearch/hermes-agent