---
title: Marco-o1: Towards Open Reasoning Models for Open-Ended Solutions
url: https://www.ml-quant.com/papers/arxiv/2411.14405/
site: ML-Quant (https://www.ml-quant.com)
updated: 2026-09-26
license: Summaries CC BY 4.0; links go to the original sources
index: https://www.ml-quant.com/llms.txt
identifier: arXiv:2411.14405
source_url: https://arxiv.org/abs/2411.14405
featured: 2024-11-27
citations: 156
topic: Other
---


# Marco-o1: Towards Open Reasoning Models for Open-Ended Solutions

The research investigates the OpenAI o1 model's ability, enhanced by Chain-of-Thought fine-tuning and innovative reasoning strategies, to adapt to wider domains lacking clear standards.

- Source: https://arxiv.org/abs/2411.14405
- Identifier: arXiv:2411.14405
- Released: 2024-11-21
- First featured: Quant Letter No. 76 (2024-11-27): https://www.ml-quant.com/issues/2024-11-27/
- Citations (Semantic Scholar): 156
- Published in: not yet
- Topic: Other

## Related

- [NV-Retriever: Improving text embedding models with effective hard-negative mining](https://www.ml-quant.com/papers/arxiv/2407.15831/): Hard-Negative Mining: The article suggests positive-aware mining methods for fine-tuning text embedding models, with the NV-Retriever-v1 model outperforming previous methods in the MTEB Retrieval benchmark.
- [LoRA-X: Bridging Foundation Models with Training-Free Cross-Model Adaptation](https://www.ml-quant.com/papers/arxiv/2501.16559/): Model Adaptation: LoRA-X enables the transfer of fine-tuning parameters across different models, enhancing the efficiency of text-to-image generation without needing original training data.
- [Dataset Size Recovery from LoRA Weights](https://www.ml-quant.com/papers/arxiv/2406.19395/): A new task called dataset size recovery aims to determine the number of samples used to train a model, with a method called DSiRe proposed for this purpose.
- [Exploring Foundation Models for Synthetic Medical Imaging: A Study on Chest X-Rays and Fine-Tuning Techniques](https://www.ml-quant.com/papers/arxiv/2409.04424/): The study investigates the use of foundation models in creating realistic medical images, specifically chest x-rays, and shows performance improvement with fine-tuning.
- [Stanfords s1 vs. DeepSeek-R1](https://www.ml-quant.com/papers/ssrn/5130864/): The s1 model, trained on a compact dataset, is cost-efficient and accurate in complex reasoning tasks, with a mechanism that allows controllable test-time scaling.
- [Instruction Finetuning Llama3-8B Model Using LoRA for Financial Named Entity Recognition](https://www.ml-quant.com/papers/arxiv/2601.10043/): The paper shows that using instruction fine-tuning and Low-Rank Adaptation with Meta's Llama 3 enhances financial named-entity recognition, leading to top performance in converting unformatted reports into organized knowledge.