Ebad Shabbir

large language model (3) model quantization (1) contrastive learning (1) factual accuracy (1) multi-modal learning (1) deepfake detection (1) model safety (1) modality alignment (1) energy efficiency (1) model pruning (1) value alignment (1) safety evaluation (1) edge computing (1) retrieval-augmented generation (1) chain-of-thought prompting (1) speculative decoding (1) parameter-efficient finetuning (1) benchmark framework (1) annotation quality (1) exploration-exploitation trade-off (1)