Skip to content
aiqbench takeAI Products

aiqbench analysis

Onton's release of Ontology 1 marks a significant shift in e-commerce search by integrating neurosymbolic reasoning to validate product attributes against objective properties rather than relying solely on seller-provided labels. This approach directly addresses the limitations of current retrieval systems when handling complex, conversational queries involving negation or specific environmental constraints like "pet-friendly." The model demonstrates that incorporating symbolic logic into neural architectures can substantially improve precision over standard large-scale search engines without requiring full catalog indexing.
Source coveringMarkTechPost
What this isA model's reading of the article and related coverage, written for aiqbench. Not the publisher's words.

Onton Releases Ontology 1: A Neurosymbolic Search Model That is 2.7x More Accurate than the World’s Best E-commerce Search Engines

MarkTechPost5 min read

Onton, a San Francisco-based search and discovery company, has released Ontology 1, a neurosymbolic model for complex, conversational, multimodal product search. On a 90-query benchmark scored by three independent LLM judges, Ontology 1 reached a mean precision@10 of 0.630, against 0.543 for Google Shopping and 0.469 for Amazon. It did this while indexing roughly 1% […] The post Onton Releases Ontology 1: A Neurosymbolic Search Model That is 2.7x More Accurate than the World’s Best E-commerce Search Engines appeared first on MarkTechPost.

Onton, a San Francisco-based search and discovery company, has released Ontology 1, a neurosymbolic model for complex, conversational, multimodal product search. On a 90-query benchmark scored by three independent LLM judges, Ontology 1 reached a mean precision@10 of 0.630, against 0.543 for Google Shopping and 0.469 for Amazon. It did this while indexing roughly 1% of their catalogs. Yes, but not as weights you download. Ontology 1 is live for end users at Onton.com, and Onton says partner access is granted case…

Excerpt shown under the publisher's RSS terms · 82 words · full text and figures are on the publisher's site

Related coverage