Paper Detail
Mind Lab, :, Vin Bo, Asher Cai, Jingwei Cao, Song Cao, Vic Cao, Amelia Chen, Andrew Chen, Kaijie Chen, Cleon Cheng, Steven Chiang, Kaixuan Fan, Hera Feng, Huan Feng, Arthur Fu, Jun Gao, Pyke Han, Nolan Ho, Ori Hong, Hailee Hou, Piers Hua, Charles Huang, Miles Jiang, Nora Jiang, Yuyi Jiang, Qiuyu Jin, Fancy Kong, Kuss Koo, Jaron Lee, Andrew Lei, Alexy Li, Dawn Li, Lucian Li, Ray Li, Ricardo Li, Smith Li, Theo Li, Allen Lin, Elliot Lin, Fan Lin, Chen Ling, Kairus Liu, Kieran Liu, Logan Liu, Neo Liu, Xiang Liu, Yuxin Lu, Maeve Luo, Pony Ma, Verity Niu, Cole Qiao, Guian Qiu, Vince Qu, Sentry, Niko Song, Vincent Wang, Bo Wu, Rio Yang, Evelyn Ye, Fiona Ye, Ina Ye, Regis Ye, Josh Ying, Atlas Zeng, Danney Zeng, Salmon Zhan, Anya Zhang, Di Zhang, Mia Zhang, Sueky Zhang, Wei Zhao, Ada Zhou, Adrian Zhou, Yuhua Zhou, Juno Zhu, Murphy Zhuang
Macaron-V1 is an open agent-model family for experiential intelligence: learning from experience in real environments and continuing to learn after deployment. It is organized around two system goals. Adaptation is pursued through recursive improvement of versioned model-harness pairs, where experience from one configuration is evaluated under an external contract and used to construct its successor. Collaboration is pursued via the Mixture-of-LoRA (MoL) architecture that freezes a base model, composes specialist LoRA adapters, and selects one LoRA per user turn. The flagship Macaron-V1-Venti combines a 744B GLM-5.2 base with four LoRAs for chat, agent, coding, and GenUI; the Qwen3.6-based Macaron-V1-Tall (50B) uses the same design for local deployment. This report presents Macaron-V1 as a co-designed system spanning architecture, algorithms, and infrastructure. The MoL architecture supports continual learning through extensible LoRA specialists. The algorithm combines Model-Harness Co-design and recursive self-improvement loop, including the UI4A component-native GenUI harness, a stateful action substrate, versioned HCP contract, and the agentic RL framework MindForge. The supporting infrastructure includes the post-training platform MinT, the long-context RL method LongStraw, and stability techniques for sparse MoE and DSA base models. We evaluate Macaron-V1 on Personal Intelligence, GenUI, and general capability benchmarks against frontier baselines. Our results validate the current system, while compounding gains from continual learning and collective intelligence remain open questions.
No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.
No ranking explanation is available yet.
No tags.
@article{lab2026macaron,
title = {Macaron-V1: Towards Open Continual Learning with Self-Improvement and Mixture-of-LoRA},
author = {Mind Lab and : and Vin Bo and Asher Cai and Jingwei Cao and Song Cao and Vic Cao and Amelia Chen and Andrew Chen and Kaijie Chen and Cleon Cheng and Steven Chiang and Kaixuan Fan and Hera Feng and Huan Feng and Arthur Fu and Jun Gao and Pyke Han and Nolan Ho and Ori Hong and Hailee Hou and Piers Hua and Charles Huang and Miles Jiang and Nora Jiang and Yuyi Jiang and Qiuyu Jin and Fancy Kong and Kuss Koo and Jaron Lee and Andrew Lei and Alexy Li and Dawn Li and Lucian Li and Ray Li and Ricardo Li and Smith Li and Theo Li and Allen Lin and Elliot Lin and Fan Lin and Chen Ling and Kairus Liu and Kieran Liu and Logan Liu and Neo Liu and Xiang Liu and Yuxin Lu and Maeve Luo and Pony Ma and Verity Niu and Cole Qiao and Guian Qiu and Vince Qu and Sentry and Niko Song and Vincent Wang and Bo Wu and Rio Yang and Evelyn Ye and Fiona Ye and Ina Ye and Regis Ye and Josh Ying and Atlas Zeng and Danney Zeng and Salmon Zhan and Anya Zhang and Di Zhang and Mia Zhang and Sueky Zhang and Wei Zhao and Ada Zhou and Adrian Zhou and Yuhua Zhou and Juno Zhu and Murphy Zhuang},
year = {2026},
abstract = {Macaron-V1 is an open agent-model family for experiential intelligence: learning from experience in real environments and continuing to learn after deployment. It is organized around two system goals. Adaptation is pursued through recursive improvement of versioned model-harness pairs, where experience from one configuration is evaluated under an external contract and used to construct its successor. Collaboration is pursued via the Mixture-of-LoRA (MoL) architecture that freezes a base model, c},
url = {https://arxiv.org/abs/2608.09819},
keywords = {cs.LG, cs.CL, huggingface daily},
eprint = {2608.09819},
archiveprefix = {arXiv},
}
{}