A web app that uses Ollama for an RLM influenced implementation. Whereas RLM was primarily concerned about context rot this leverages the chunking approach to work around the small context windows of local LLMS.
From the MIT RLM white paper:
rom the paper:
"Even without sub-calling capabilities, our ablation of the RLM is able to scale beyond the context limit of the model, and outperform the base model and other task-agnostic baselines on most long context settings."
This approach keeps the spirit of this.
All state is stored in the browser in indexed DB including text extracted from PDF and Docs, and the App will attempt to detect a local running Ollama instance.
JAL_UK•1w ago
From the MIT RLM white paper:
rom the paper:
"Even without sub-calling capabilities, our ablation of the RLM is able to scale beyond the context limit of the model, and outperform the base model and other task-agnostic baselines on most long context settings."
This approach keeps the spirit of this.
All state is stored in the browser in indexed DB including text extracted from PDF and Docs, and the App will attempt to detect a local running Ollama instance.