BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//AMSE//Event Calendar//FR
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
UID:event-11788@www.amse-aixmarseille.fr
DTSTAMP:20260414T114238Z
CREATED:20260414T114238Z
LAST-MODIFIED:20260414T114238Z
STATUS:CONFIRMED
SEQUENCE:0
SUMMARY:amse seminar - Marc Ratkovic
DTSTART:20250324T103000Z
DTEND:20250324T114500Z
DESCRIPTION:Text data pose significant challenges for statistical inference
  due to their high dimensionality and unstructured nature\, and conventiona
 l methods based on fixed embeddings or topic models often oversimplify ling
 uistic complexity without providing formal inferential guarantees. In this 
 work\, I integrate large language models (LLMs) with statistical inference 
 by employing them to generate latent contexts that serve as auxiliary infor
 mation. I introduce a clause function that quantifies the interaction betwe
 en an observed text string and its LLM-generated latent contexts\, and by a
 veraging over these contexts\, obtain string-level statistics. Under standa
 rd support and ignorability assumptions along with additional regularity co
 nditions\, I characterize the limiting distribution of these statistics—a
  result that extends to averages\, regression coefficients\, and robust ran
 k- and quantile-based alternatives. Applications to a two-sample test and a
  regression model with text-based predictors and outcomes demonstrate the u
 tility of this framework for rigorous inference on textual data.\\n\\nConta
 ct: Nicolas Clootens : nicolas.clootens[at]univ-amu.frRomain Ferrali : roma
 in.ferrali[at]univ-amu.fr\n\nPlus d'informations: https://www.amse-aixmarse
 ille.fr/fr/evenements/marc-ratkovic
LOCATION:Îlot Bernard du Bois - Amphithéâtre\, AMU - AMSE\, 5-9 boulevar
 d Maurice Bourdet\, 13001 Marseille
URL;VALUE=URI:https://www.amse-aixmarseille.fr/fr/evenements/marc-ratkovic
CONTACT:Nicolas Clootens : nicolas.clootens[at]univ-amu.frRomain Ferrali : 
 romain.ferrali[at]univ-amu.fr
TRANSP:OPAQUE
END:VEVENT
END:VCALENDAR