MRKL Systems: A modular, neuro-symbolic architecture that combines large language models, external knowledge sources and discrete reasoning

Abstract

Huge language models (LMs) have ushered in a new era for AI, serving as a gateway to natural-language-based knowledge tasks. Although an essential element of modern AI, LMs are also inherently limited in a number of ways. We discuss these limitations and how they can be avoided by adopting a systems approach. Conceptualizing the challenge as one that involves knowledge and reasoning in addition to linguistic processing, we define a flexible architecture with multiple neural models, complemented by discrete knowledge and reasoning modules. We describe this neuro-symbolic architecture, dubbed the Modular Reasoning, Knowledge and Language (MRKL, pronounced "miracle") system, some of the technical challenges in implementing it, and Jurassic-X, AI21 Labs' MRKL system implementation.

LAVE: LLM-Powered Agent Assistance and Language Augmentation for Video Editing

“…To collaborate with the agent, users begin the process by typing their editing objectives. The agent interprets the user's objectives and formulates an action plan to fulfill them [31,60,62,81]. The agent operates in two modes: Planning and Executing.…”

Genomic Surveillance for SARS-CoV-2 — China, September 26, 2022 to January 29, 2023

“…This information is then passed to an external symbolic planner, which efficiently determines the optimal sequence of actions from the current state to the target state. MRKL [71] is a modular, neural-symbolic AI architecture, where LLMs process the input texts, routes them to each of the experts, and then pass them through the LLMs' outputs. CO-LLM [156] consider that LLMs are good at generating high-level plans, but not good at low-level controlling.…”

“…ChemCrow [8] presents an LLM-based chemical agent aimed at accomplishing tasks in the fields of organic synthesis, drug discovery, and material design, with the help of seventeen expert-designed tools. MRKL Systems [71], OpenAGI [51] incorporates various expert systems such as knowledge bases and planners, invoking them to access domain-specific information in a systematic manner. (3) Language Models.…”

Can Foundation Models Wrangle Your Data?

“…Understanding the tradeoffs around training a specialized FM versus utilizing an existing off-the-shelf FM is a critical research question. Beyond academic efforts, AI21 Labs recently released MRKL-a system for augmenting FMs with external knowledge sources and symbolic reasoning experts-demonstrating how FMs can be utilized to ingest, reason and operate over varied, domain-specific data stores [39].…”