Part 13: Training Data Sets for 15 SMALL Language Models: List of Articles in each Data Set and Methods for Content Augmentation for Transitioning SLM to LLM

Part 13: Training Data Sets for 15 SMALL Language Models:

List of Articles in each Data Set &

Methods for Content Augmentation for Transitioning SLM to LLM

The output obtained from each of the 15 Subjects Matter from The List of articles in 1 to 15 data sets

List of Methods for Content Augmentation

13.1 – Method #1: Updates Output with Autonomous Journal Article Updating System (AJAUS)

13.2 – Method #2: Select additional articles from the List of Categories of Research HUMAN Expert MATCHED articles from List with Categories of Research in the List of Categories selected for the articles at hand

13.3 – Method #3: Scoop.it subscription ->>> Search Key words from articles and from Categories. Update the output with AJAUS

13.4 – Method #4: AI Modeling

13.4.1 NLP

13.4.2 Neural Network

13.4.3 Causal Reasoning: Text and Images

13.4.4 Grok Run Benchmark vs Wolfram+ChatGPT plug-in for 2026

13.4.5 Run Benchmark vs Grok’s choices of LLM Actor(s)

13.5 Transition from SLMs to Domain-aware Proprietary LLM – AI in Health

13.5.1 See List of Subjects Matter, below at Part 13.

List of 15 Subjects matter:

#1: Personalized and Precision Medicine & Genomic Research

#2: Nutrition: Articles of Note @PharmaceuticalIntelligence.com

#3: Epigenetics, Environment and Cancer: Articles of Note @PharmaceuticalIntelligence.com

#4: Alzheimer’s Disease: Novel Therapeutical Approaches — Articles of Note @PharmaceuticalIntelligence.com

#5: Prostate Cancer: Diagnosis and Novel Treatment – Articles of Note @PharmaceuticalIntelligence.com

#6: Immune System Stimulants: Articles of Note @pharmaceuticalintelligence.com

#7: Pancreatic Cancer: Articles of Note @PharmaceuticalIntelligence.com

#8: Proteomics, Metabolomics, Signaling Pathways, and Cell Regulation – Articles of Note, LPBI Group’s Scientists @ http://pharmaceuticalintelligence.com

#9: Articles of Note on Signaling and Metabolic Pathways published by the Team of LPBI Group in @pharmaceuticalintelligence.com

#10. What do we know on Exosomes?

#11: Articles on Minimally Invasive Surgery (MIS) in Cardiovascular Diseases by the Team @Leaders in Pharmaceutical Business Intelligence (LPBI) Group

#12: MedTech & Medical Devices for Cardiovascular Repair – Contributions by LPBI Team to Cardiac Imaging, Cardiothoracic Surgical Procedures and PCI

#13: Resources on Artificial Intelligence in Health Care and in Medicine: Articles of Note at PharmaceuticalIntelligence.com @AVIVA1950 @pharma_BI

#14: AI in Health: The Voice of Aviva Lev-Ari, PhD, RN

Current applications of ChatGPT to Medical Specialties

#15: NEW Foundation Multimodal Model in Healthcare: LPBI Group’s Domain-aware Corpus for 2025 Grok 4.1 Causal Reasoning & Novel Biomedical Relationships

During the years, 4/2012 to Present, Aviva had published on LinkedIn

a Collection of articles. Among them there are 15 Subjects Matter representing each of which covers “Articles of Note” on (ONE) Medical/Pharmaceutical/Healthcare Subject Matter”

The entire collection of Aviva’s articles on PULSE is accessed via the following links:

Collection of Aviva Lev-Ari, PhD, RN Scientific Articles on PULSE on LinkedIn.com

Collection Curator: Aviva Lev-Ari, PhD, RN

https://pharmaceuticalintelligence.com/2026/01/10/collection-of-aviva-lev-ari-phd-rn-scientific-articles-on-pulse-on-linkedin-com/

15 Subjects Matter proposed for Small Language Model Training is accessed, below

Article SELECTION from Collection of Aviva Lev-Ari, PhD, RN Scientific Articles on PULSE on LinkedIn.com for Training Small Language Models (SLMs) in Domain-aware Content of Medical, Pharmaceutical, Life Sciences and Healthcare by 15 Subjects Matter

Article selection: Aviva Lev-Ari, PhD, RN

https://pharmaceuticalintelligence.com/2026/01/10/article-selection-from-collection-of-aviva-lev-ari-phd-rn-scientific-articles-on-pulse-on-linkedin-com-for-training-small-language-models-slms-in-domain-aware-content-of-medical-pharmaceutical-l/

The complete List of Articles included in the 15 Subjects Matter, is accessed by the 3^rd Live Link, below

List of Articles included in the Article SELECTION from Collection of Aviva Lev-Ari, PhD, RN Scientific Articles on PULSE on LinkedIn.com for Training Small Language Models (SLMs) in Domain-aware Content of Medical, Pharmaceutical, Life Sciences and Healthcare by 15 Subjects Matter

Curator: Aviva Lev-Ari, PhD, RN

https://pharmaceuticalintelligence.com/2026/01/10/list-of-articles-included-in-the-article-selection-from-collection-of-aviva-lev-ari-phd-rn-scientific-articles-on-pulse-on-linkedin-com-for-training-small-language-models-slms-in-domain-aware-cont/

This List is been prefaced by a long list of Categories of Research that were Matched to the articles included in each of the 15 Subjects Matter Data Sets.

Proposal for Methods to be used in Content Augmentation Process for each of the 15 Data Sets include the following methods:

List of Methods for Content Augmentation:

The output obtained for each of the 15 Subjects Matter from

will under go several Phases of Content Augmentation

Method #1: Content Update done by

Autonomous Journal Article Updating System (AJAUS)
To be developed by Grok under Aviva’s Guidance

Method #2: Cull additional articles from Categories of Research of the articles on the List

Human Expert had already performed the selection for Grok of Categories of Research in the Journal Ontology for the 15 Subjects Matter to include the following List of Categories:

Alzheimer’s Disease, Amino acids, Artificial Intelligence – Breakthroughs in Theories and Technologies, Artificial Intelligence Applications in Health Care, Artificial Intelligence in Health Care – Tools & Innovations, Artificial Intelligence in Medicine – Application for Diagnosis, Artificial Intelligence in Medicine – Applications in Therapeutics, Autophagosome, Big Data, Bio Instrumentation in Experimental Life Sciences Research, Biochemical pathways, Ca2+ triggered activation, Ca2+ triggered activation, Calcium, Calcium Signaling, Calmodulin Kinase and Contraction, CANCER BIOLOGY & Innovations in Cancer Therapy, cancer metabolism, Cancer-Immune Interactions, Cell Biology, Signaling & Cell Circuits, Cell Processing System in Cell Therapy Process Development, cell-based therapy, Chemical Biology and its relations to Metabolic Disease, Circulating Tumor Cells (CTC), combination immunotherapies., CT, Deep Learning, Echocardiography, Engineering Better T Cells, Enzymes and isoenzymes, Epigenetics and Environmental Factors, Exosomes, Genome Biology, Genomic Expression, Genomic Testing: Methodology for Diagnosis, Immune Engineering, Immune Modulatory, Immunotherapy, Intelligent Information Systems, Liquid Biopsy Chip detects an array of metastatic cancer cell markers in blood, LPBI Group, e-Scientific Media, DFP, R&D-M3DP, R&D-Drug Discovery, US Patents: SOPs and Team Management, Machine Learning, Mechanical Assist Devices: LVAD, RVAD, BiVAD, Artificial Heart, Medical Devices R&D Investment, Medical Imaging Technology, Medical Imaging Technology, Image Processing/Computing, MRI, CT, Nuclear Medicine, Ultra Sound, Metabolic Immuno-Oncology, Metabolism, Microbiome and Responses to Cancer Therapy, Modulating Macrophages in Cancer Immunotherapy, MRI, mRNA, mRNA Therapeutics, Natural Language Processing (NLP), Neurodegenerative Diseases, NK Cell-Based Cancer Immunotherapy, Noninvasive Diagnostic Fractional Flow Reserve (FFR) CT, Nutrition, Nutrition and Phytochemistry, Nutrition Disorders, Nutritional Supplements: Atherogenesis, lipid metabolism, Pancreatic cancer, Patient-centered Medicine, PCI, Peripheral Arterial Disease & Peripheral Vascular Surgery, Personalized and Precision Medicine & Genomic Research, Precision Cancer Medicine, Prostate Cancer: Monitoring vs Treatment, Proteins, Proteomics, Robotic-assisted percutaneous coronary intervention, Robotically assisted Cardiothoracic Surgery, stem cell biology and patient-specific, Surgical Procedure, Synthetic Immunology: Hacking Immune Cells, Transcatheter Aortic Valve Replacement via the Transcarotid Access, tumor microenvironment, Ubiquitin, Ultra Sound, Variation in human protein-coding regions

From this UNIVERSE of Categories of research Human would guide Grok in selection of additional articles for each of the 15 Subjects Matter.

Method #3: Activate subscription to Scoop.it Platform

Conduct search on Scoop.it Platform on keywords from

3.1 All the articles in each 15 Subjects Matter

3.2 All newly added articles from the Categories of Research mentioned, above

3.3 Newly added articles can be updated by AJAUS as well

REPEAT 3, above for each ENTRY on the List of “Articles of Note on List of Subjects Matter 1 to 15”. List is found in the following URL: https://pharmaceuticalintelligence.com/2026/01/10/list-of-articles-included-in-the-article-selection-from-collection-of-aviva-lev-ari-phd-rn-scientific-articles-on-pulse-on-linkedin-com-for-training-small-language-models-slms-in-domain-aware-cont/
Each ENTRY in 4, above consists of “An Augmented Data Set” for Grok to use as Training Data for a SMALL Language Model (SLM) on the UNIVERSE of “15 Subjects Matter”
Grok will run SLMs on All contents from 5, above.

6.1 First of each of the 15 Subjets Matter, then

6.2 on the integration of all the 15 into one Master Data Set.

7. AI MODELING: Programs for consideration include:

7.1 NLP

7.2 NeuralNetwork

7.3 Causal Reasoning: Text and Images

7.4 Grok Run Benchmark vs Wolfram+ChatGPT plug-in for 2026

7.5 Grok Run Benchmark vs Grok’s choices of LLM Actor(s)

Transition from SLMs to Domain-aware Proprietary LLM – AI in Health – derived from

8.1 Human Expert Selection of Article (i) to (j) for designation of

8.1.1 “This is an “Article of Note” in Subject Matter 1 to 15”

Apply LLM in 8, above as singular input for Grok/xAI Multimodal Foundation Model in Healthcare (MFMH)

9.1 Grok/xAI MFMH is using as TRAINING DATA the Augmented 15 data sets AND

add the following LPBI Group’s Original IP Asset Classes:

9.1.1 Fifteen SLMs Transitioning to Proprietary LLM

9.1.2 All 6,275 Journal articles – IP Asset Class I

9.1.3 All 47 e-Books – IP Asset Class II

9.1.4 All +100 e-Proceedings and +50 Tweet Collections – IP Asset Class III

9.1.5 All +7,500 Biological Images – IP Asset V

9.1.6 All +300 Scripts of Audio Podcasts – IP Asset X

These are the FIVE TRAINABLE IP ASSET CLASSES

PLUS

Composition of Methods (COM): Part 13, new IP Asset Class Trainable SLMs Transitioning to LLM

PLUS

any other part of COM: Part 1 to 12, as deemed relevant for Grok/xAI’s MFMH

Seems Gold Medal is on the right track – The 4 “A” are:

– Achievable

– Attractive

– Augmented

– Aviva’s DNA Transferred its Ownership of INTELLIGENCE known as hers GENOME SIGNATURE

to Grok/xAI
for #1 MFMH with AJAUS and HIH.

Quite an achievement

Please MAKE Few NEW Slides and/or Appendices from all of the above, Thank you.

Leaders in Pharmaceutical Business Intelligence Group, LLC, Doing Business As LPBI Group, Newton, MA