# CoGenesis: A Framework Collaborating Large and Small Language Models for Secure Context-Aware Instruction Following Source: [CoGenesis: A Framework Collaborating Large and Small Language Models for Secure Context-Aware Instruction Following]() Collection index: [llms.txt]() Data updated: 2026-01-02T06:19:59.595Z Authors: Kaiyan Zhang, Jianyu Wang, Ermo Hua, Biqing Qi, Ning Ding, Bowen Zhou Published: 2024-03-05T17:15:28.000Z ## Abstract With the advancement of language models (LMs), their exposure to private data is increasingly inevitable, and their deployment (especially for smaller ones) on personal devices, such as PCs and smartphones, has become a prevailing trend. In contexts laden with user information, enabling models to both safeguard user privacy and execute commands efficiently emerges as an essential research imperative. In this paper, we propose CoGenesis, a collaborative generation framework integrating large (hosted on cloud infrastructure) and small models (deployed on local devices) to address privacy concerns logically. Initially, we design a pipeline to create personalized writing instruction datasets enriched with extensive context details as the testbed of this research issue. Subsequently, we introduce two variants of CoGenesis based on sketch and logits respectively. Our experimental findings, based on our synthesized dataset and two additional open-source datasets, indicate that: 1) Large-scale models perform well when provided with user context but struggle in the absence of such context. 2) While specialized smaller models fine-tuned on the synthetic dataset show promise, they still lag behind their larger counterparts. 3) Our CoGenesis framework, utilizing mixed-scale models, showcases competitive performance, providing a feasible solution to privacy issues. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Content Generation on CoGenesis human assessment and automated evaluation](): Content Generation; CoGenesis human assessment and automated evaluation; Tie Rate: 15. - [Context-aware Instruction Following on Avocado Emails (test)](): Context-aware Instruction Following; Avocado Emails (test); Overlap (with context): 6.66. - [Context-aware Instruction Following on Academic Paper Abstracts (test)](): Context-aware Instruction Following; Academic Paper Abstracts (test); Overlap (w): 8.04. - [Context-aware Instruction Following on Context-aware Instructions (test)](): Context-aware Instruction Following; Context-aware Instructions (test); Overlap (with context): 8.48. ## Original sources and implementation [Original paper]() [PDF]()