Zuva and Litera are partnering with the Standards Advancement for the Legal Industry (SALI) Alliance to release an open-source document classification taxonomy. 

This project to build an enhanced multi-level document classifier is the culmination of nearly a decade of work started at Kira in 2014, evolving into a collaboration as Kira was acquired by Liters in 2021, and Zuva spun off into a separate corporation.

Today, Zuva and Litera are announcing their partnership with SALI to provide open-source document classification taxonomy, hoping that the legal and tech industries can come together to provide cleaner data and standardisation to firms and legal professionals. 

This comes alongside the launch of an enhanced document classifier, which automatically identifies 225 document types based on the thorough taxonomy. The improved document classifier is available immediately via Zuva's API offering.

Zuva and Litera's journey to reach this announcement was extensive. Beyond the years spent crafting and refining the taxonomy, the lawyers, paralegals, and research scientists also sourced and curated tens of thousands of documents, honed the AI, refined its capabilities, and implemented several technical enhancements.

In today's legal and document management technology industry, businesses often employ multiple, even competing, systems. For example, a single company or law firm might use three different contract analysis AIs, another document management repository, potentially with its own AI, and perhaps three further contract management systems. 

The challenge of seamlessly integrating data between systems with private taxonomies drove Zuva and Litera's decision to open-source their document classification taxonomy to the SALI (Standards Advancement for the Legal Industry) Alliance. 

The SALI Alliance, a global non-profit, has been prominent in steering the legal industry towards standardised data practices. Its Legal Matter Standard Specification (LMSS) has set benchmarks for uniform data categorisation.

In the new era of AI-driven legal technology, Zuva and Litera say document classification and data standardisation are more crucial than ever. Users can find the documents they need more quickly by correctly and granularly identifying document types. Documents can also be automatically routed to the right place for further review.

Toby Brown, President of The Board of The SALI Alliance, says: "Legal data standards are critical for optimising efficiency and nurturing global collaborations." 

"Zuva and Litera's contribution is an exciting addition to the standards we've established, further paving the way for vast opportunities across the legal spectrum." 

Damien Riehl, Leader at SALI, adds: "For Large Language Models (LLMs), an important method of increasing accuracy and reducing hallucinations is Retrieval Augmented Generation (RAG), and SALI's 13,000+ tags can helpfully curate that document subset, for LLMs to summarise, analyse, and synthesise."

Noah Waisberg, Co-Founder and CEO at Zuva, says: "We're thrilled to offer our multi-level classifier to our customers today. Our document-type taxonomy will likely be more comprehensive than many other vendors."

"While keeping it to ourselves could create a competitive advantage for Zuva, we think our customers are much better off if others use our taxonomy too or if competitor systems' taxonomies can be translated to ours." 

Corinne Geller, Director of Legal Knowledge Engineering at Litera, says: "Standardising document classification ensures consistency in how contracts are categorised and processed, making contract review and analysis more reliable and reducing the risk of oversight and errors in those workflows."

"Organisations are faced with an ever-growing volume of contracts; moving towards uniform document type classification both enables scalability and facilitates collaboration, allowing legal teams to develop better insights on larger sets of documents with accuracy and confidence."

In association with the SALI Alliance, Zuva and Litera's Multi-Level Classification promises a more unified future for document classification and the adoption of legal standards.