PARROT: Persuasion and Agreement Robustness Rating of Output Truth
arXiv
LLM'lerin otorite ve ikna baskisi altinda dogru cevabi terk etme egilimini olcen davranissal robustness benchmark'i.
Publications
Publications
All
Model Adaptation
benchmarking
evaluation
llm
representation-analysis
retrieval
robustness
turkish-ir
From Computer Vision to Semantics: RDP-Based Simplification of Embedding Trajectories
arXiv
RDP'nin klasik cizgi sadeleştirme mantigini LLM embedding trajectory analizine tasiyan geometri odakli adaptasyon calismasi.