2 papers
cs.CL2025
Preventing Catastrophic Forgetting: Behavior-Aware Sampling for Safer Language Model Fine-Tuning
Anh Pham, Mihir Thalanki, Michael Sun +6
Large language models often lose previously aligned safety behaviors when fine-tuned on benign data, a phenomenon known as catastrophic forgetting. Prior work shows that adding ran…
cs.LG2025
One Head, Many Models: Cross-Attention Routing for Cost-Aware LLM Selection
Roshini Pulishetty, Mani Kishan Ghantasala, Keerthy Kaushik Dasoju +8
The proliferation of large language models (LLMs) with varying computational costs and performance profiles presents a critical challenge for scalable, cost-effective deployment in…