Seraphina Goldfarb-Tarrant
Scholar

Seraphina Goldfarb-Tarrant

Google Scholar ID: ywWpui8AAAAJ
University of Edinburgh
Natural Language ProcessingArtificial Intelligence
Homepage↗Google Scholar↗
Citations & Impact
All-time
Citations
677
 
H-index
11
 
i10-index
11
 
Publications
20
 
Co-authors
7
list available
Contact
No contact links provided.
Publications
8 items
Learning is Forgetting: LLM Training As Lossy Compression
2026
Cited
0
Lost in Simulation: LLM-Simulated Users are Unreliable Proxies for Human Users in Agentic Evaluations
2026
Cited
0
A Good Plan is Hard to Find: Aligning Models with Preferences is Misaligned with What Helps Users
2025
Cited
0
Scalpel vs. Hammer: GRPO Amplifies Existing Capabilities, SFT Replaces Them
2025
Cited
0
The Multilingual Divide and Its Impact on Global AI Safety
2025
Cited
0
MAPS: A Multilingual Benchmark for Global Agent Performance and Security
2025
Cited
0
Safer or Luckier? LLMs as Safety Evaluators Are Not Robust to Artifacts
2025
Cited
0
Who Does the Giant Number Pile Like Best: Analyzing Fairness in Hiring Contexts
2025
Cited
0
Resume (English only)
Co-authors
7 total
Nanyun (Violet) Peng
Nanyun (Violet) Peng
Associate Professor, UCLA
Co-author 2
Co-author 2
Co-author 3
Co-author 3
Tuhin Chakrabarty
Tuhin Chakrabarty
Stony Brook University , Department of Computer Science
Co-author 5
Co-author 5
Co-author 6
Co-author 6
Co-author 7
Co-author 7