3 papers
cs.LG2026
HeadGuard: Selective Head Protection for Low-Bit VLM KV-Cache Quantization
Nenad Banfic
Low-bit key-value (KV) cache quantization saves storage but can sharply degrade vision-language model (VLM) accuracy. We introduce HeadGuard, a composable head-protection method th…
cs.AI2026
Data Scale, Not Latency, Shapes Cross-Lingual Encoder Transfer in Streaming ASR
Nenad Banfic
Adapting a streaming speech recognition model to a new language requires choosing between two plausible warm starts: a multilingual (ML) encoder or an English-only (EN) encoder. Th…
cs.AI2026
Pushing the Limits of On-Device Streaming ASR: A Compact, High-Accuracy English Model for Low-Latency Inference
Nenad Banfic, David Fan, Kunal Vaishnavi +5
Deploying high-quality automatic speech recognition (ASR) on edge devices requires models that jointly optimize accuracy, latency, and memory footprint while operating entirely on…