Puria Radmard
Scholar

Puria Radmard

Google Scholar ID: O3Lza0oAAAAJ
Univeristy of Cambridge
Google Scholar↗
Citations & Impact
All-time
Citations
47
 
H-index
3
 
i10-index
1
 
Publications
6
 
Co-authors
3
list available
Contact
No contact links provided.
Publications
11 items
Inoculation Midtraining with Learned Neologisms
2026
Cited
0
Constitutional Midtraining: Content Presence Drives Alignment Gains
2026
Cited
0
Self-Generated Text Recognition: Quality Heuristics, Cross-Task Transfer, and Downstream Bias in LLM Evaluation
2026
Cited
0
A transformer architecture alteration to incentivise externalised reasoning
2026
Cited
0
Diagnosing Pathological Chain-of-Thought in Reasoning Models
2026
Cited
0
Chain-of-thought obfuscation learned from output supervision can generalise to unseen tasks
2026
Cited
0
Alignment Pretraining: AI Discourse Causes Self-Fulfilling (Mis)alignment
2026
Cited
2
Setting up for failure: automatic discovery of the neural mechanisms of cognitive errors
2025
Cited
0
Resume (English only)
Co-authors
3 total
Yassir Fathullah
Yassir Fathullah
Google DeepMind
Aldo Lipani
Aldo Lipani
Professor (Associate) at UCL
Máté Lengyel
Máté Lengyel
Professor of Computational Neuroscience, University of Cambridge; Senior Research Fellow, CEU