US 12,271,704 B2
Detection of veracity of responses in machine comprehension question and answer models
Kunal Sawarkar, Franklin Park, NJ (US); and Shivam Raj Solanki, Austin, TX (US)
Assigned to INTERNATIONAL BUSINESS MACHINES CORPORATION, Armonk, NY (US)
Filed by International Business Machines Corporation, Armonk, NY (US)
Filed on Sep. 21, 2022, as Appl. No. 17/934,029.
Prior Publication US 2024/0095458 A1, Mar. 21, 2024
Int. Cl. G06F 40/30 (2020.01); G06F 40/166 (2020.01); G06F 40/279 (2020.01); G06F 40/35 (2020.01); G06F 40/40 (2020.01); G06F 40/20 (2020.01)
CPC G06F 40/35 (2020.01) [G06F 40/166 (2020.01); G06F 40/279 (2020.01); G06F 40/40 (2020.01); G06F 40/20 (2020.01); G06F 40/30 (2020.01)] 20 Claims
OG exemplary drawing
 
1. A system, comprising:
a memory that stores computer executable components;
a processor that executes at least one of the computer executable components that:
receives a query;
alters a text corpus to produce one or more altered text corpora, wherein the one or more altered text corpora maintain semantic integrity of the text corpus and respectively introduce different modifications to the text corpus;
generates a first answer to the query based on the text corpus;
generates one or more additional answers based on the one or more altered text corpora;
compares the first answer with the one or more additional answers;
generates a respective similarity score for each comparison of the first answer with an additional answer of the one or more additional answers; and
generates a veracity score for the first answer based on the respective similarity scores, wherein the veracity score indicates a likelihood of the first answer being correct.