28 citations · 60 across the 12 of their papers we have counts for
4 papers · 1 filter
Does Every User Need a Private LoRA? Decoupling Personalization from Per-User Adaptation
Songyuan Sui, Srikanth Malla, Chiho Choi +1
Personalized large language models often require a complete adaptation state for each user. However, this paradigm scales poorly as the user population grows. We revisit this desig…
Steering Interference Reflects the Model's Defaults, Not the Behavior Directions
Srikanth Malla, Chiho Choi, Joon Hee Choi
Activation steering promises modular control of language model behavior: a behavior such as politeness corresponds to a direction in a model's activations, and adding that directio…
The Geometry of Refusal: Why Post-Hoc Safety Is Fragile and Pretraining-Time Safety Persists
Srikanth Malla, Chiho Choi, Joon Hee Choi
Post-hoc safety training (RLHF, DPO) is the dominant way to align large language models, yet jailbreaks (Zou et al., 2023), fine-tuning attacks (Qi et al., 2024), and activation-sp…
ADEPT: Adaptive Dynamic Early-Exit Process for Transformers
Sangmin Yoo, Srikanth Malla, Chiho Choi +2
The inference of large language models imposes significant computational workloads, often requiring the processing of billions of parameters. Although early-exit strategies have pr…