RResearcharAcademic radar for researchers

Feed Venues Groups Network Inbox Profile

Your academic radar

One personalized feed of papers, calls for papers, collaborators and datasets — plus research groups and paid-in-points study recruitment.

Create account Log in

Top venues

Browse all

Loading…

HajeriLM Research: Model Training, Representation Legibility, and LM-Head Locking · Researchar

Back to feed

Paperopenalex

OPO

HajeriLM Research: Model Training, Representation Legibility, and LM-Head Locking

Mohamed Al Hajeri · OSF Preprints (OSF Preprints) 2026 · 2026

0citations

Topic ModelingNatural Language Processing TechniquesMultimodal Machine Learning ApplicationsRepresentation (politics)Computer scienceArtificial intelligence+2

OPOOSF Preprints (OSF Preprints)Repository

Download PDF Publisher

HajeriLM Research: Model Training, Representation Legibility, and LM-Head Locking

Open Access

Mohamed Al Hajeri·OSF Preprints (OSF Preprints)(2026)

Download Open in tab

Viewing in embedded reader. If this publisher restricts inline iframe preview:

Open directly Download PDF

Citations by source

Synced 2026-09-17T02:30:12.183166Z

0OpenAlex · primary0Semantic Scholar0Crossref

Counts differ because each database indexes a different set of publications. We treat OpenAlex as the canonical count; Google Scholar is not shown (no API, and crawling it violates its ToS).

Abstract

Independent AI research by Mohamed Al Hajeri on language-model training, model merging, representation legibility, LM-head locking, gated representation fusion, and transformer behavior.

Discussion (0)

No comments yet — start the discussion below.