Why do we need arguments in RAG?
RAG (Retrieval-augmented Generation) is only as good as the retrieval. If retrieval fails, the LLM will either still make up an answer (and possibly hallucinate), or simply not be able answer the question. For example, ChatGPT (in its latest 4o version and using web search) is very good in providing reasonable answers to questions about […]
Unlocking competitive advantage with AI argument clustering
How do you best capture emerging trends before it is too late? summetix changes the landscape with its powerful argument clustering capabilities. It’s time to retire the word cloud. While it may provide a surface-level understanding, it often falls short in revealing the deeper context behind the data. Run a word cloud on customer reviews […]
Spotting Product Pipeline Issues Six Months in Advance: A Case Study
Timely identification of product issues can make the difference between leading the market and falling behind. Recently, SUMMETIX helped a leading oral health technology company anticipate product problems six months in advance, showcasing the power of advanced analytics and AI-driven insights. The Challenge Our customer, a global leader with operations in almost every country, faced […]
Blog-Beitrag: Die Herausforderungen der Neo Bank enthüllen – Einblicke in die App Store Bewertungen
Unveiling Neo Bank Challenges – Insights from App Store Reviews This blog post was originally posted on LinkedIn. Please visit the post there to engage with comments. As the fintech industry continues to revolutionize traditional banking, the emergence of Neo Banks has been met with both excitement and skepticism. These digital-first institutions promise convenience, innovation, […]
A Comparative Exploration of Weight Loss Medications
The fight against obesity has become a pressing public health issue today. Especially after multiple articles around Eli Lilly`s record-breaking factory opening in Alzey, Germany. With alarming obesity rates affecting almost half of the female and more than 60% of the male population in Germany, and a considerable proportion of the population suffering from obesity, […]
Enhancing Retrieval-Augmented Generation with Argument Mining: A Paradigm Shift in AI
Retrieval-Augmented Generation (RAG) has emerged as a game-changer, integrating the power of generative AI with an organization’s internal data. The traditional retrieval approach, which often relies on semantic search in vector databases, comes with several challenges, such as limited search context and increased storage and computational costs. At summetix, we have taken a revolutionary approach […]
Aktuelle Stellenausschreibungen
summetix hat derzeit mehrere offene Stellen, unter anderem suchen wir Fullstack-Entwickler, Vertriebsassistenten und Werkstudenten (IT und Business Support). Wenn Sie an einer Zusammenarbeit mit uns interessiert sind, kontaktieren Sie uns über jobs@summetix.com.
summetix ist Frankfurt Forward’s Startup des Monats (August)
summetix wurde von Frankfurt Forward zum Startup des Monats August 2023 gekürt. Als Projekt der Wirtschaftsförderung Frankfurt fördert Frankfurt Forward die weitere Entwicklung von Innovation und Digitalisierung in Frankfurt am Main.
Publikationsalarm: Crowdsourcing auf sensiblen Daten mit datenschutzgerechtem Text Rewriting
Die meisten Aufgaben im NLP erfordern gelabelte Daten. In diesem kürzlich veröffentlichten Beitrag (auf dem 17. Linguistic Annotation Workshop, der in Verbindung mit der ACL 2023 stattfand) untersuchen wir, wie die Entfernung persönlich identifizierbarer Informationen sowie die Anwendung von differentiellem Privacy Rewriting die Nutzung von Crowdsourcing für Texte mit datenschutzrelevanten Informationen ermöglicht.
Veranstaltungshinweis: summetix auf der diesjährigen AICon in Darmstadt
summetix wird auf der diesjährigen Hessischen AICon in Darmstadt vertreten sein! Am 6. Juli zeigen wir Ihnen, wie Sie Argument Mining und Chat-GPT kombinieren können, um Halluzinationen zu vermeiden und wie summetix KI-Technologie einsetzt, um weltweit führende Unternehmen sechs Monate früher als bisher über Produktprobleme zu informieren.