I was initially using a horizontal split view for the spreadsheet and chat in @financebrainai.
Users were struggling with this on smaller browsers, so now I'm trying out a side-by-side view.
@ChengRick4 Playing around with the completion token setting may help. I recommend reviewing this if you haven't seen it already: https://t.co/cpCQAJTamT
@levelsio The ones at MK say "excuse me" or something when you get in the way. It's pretty funny to see multiple of them get into traffic jams with people walking around.
🗺️ Plan-and-Execute Agents 🗺️
Inspired by BabyAGI and the recent Plan-and-Solve paper, we're introducing a new type of @langchain agent
We think these are better for more complex tasks, at the cost of more calls to the LLM
Blog: https://t.co/xBwXQ92cSJ
🧵
The core differences between GPT4 and 3.5 is the ability to perform complex tasks. In this post, we present a complete roadmap towards LLMs complex reasoning abilities, covering the full development stages: pretraining, SFT, RL, CoT prompting, and eval. https://t.co/1rzx02LNlx
@BeeLoudxyz@_buildspace You can manually enter any data into it and ask questions, whether it is stock prices, public company information, crypto prices, or even non-financial data. In the future, I will add integrations to pull in external data without needing to manually input it.
Finance Brain can now answer more complex finance and accounting questions with the upgrades that were released this week.
Here's a demo where I calculate the net present value of a project with multiple cash flows
@_buildspace
Plug-and-Play Compositional Reasoning for LLMs
-LLM as natural language planner infers sequence of tool use:
Python
Vision models
Search engines
Modules tailored to user
-SoTA 86.5% accuracy on ScienceQA
-GPT-4 as planner unlocks rational tool selection
https://t.co/U7WSNiU2Qe
I'm working on making it easier for first-time visitors of @financebrainai to understand how the application works.
This week I added a basic landing page, these are my initial results.
@_buildspace