7 papers
SPARK: Spatial Policy-driven Adaptive Reinforcement learning for Knowledge distillation
Mohamed Jismy Aashik Rasool, Shabir Ahmad, Gisong Oh +1
Low-bit quantization enables deployment of image restoration (IR) networks on resource-constrained devices, but introduces rounding noise that disproportionately degrades high-freq…
CUBE: Contrastive Understanding by Balanced Experiments
Dongseok Kim, Hyoungsun Choi, Mohamed Jismy Aashik Rasool +1
Post-hoc explanation depends on how model queries are organized. We propose CUBE, a design-based framework that explains a trained predictive model through balanced low--high probe…
CLAPS: Aleatoric-Epistemic Scaling via Last-Layer Laplace for Conformal Regression
Dongseok Kim, Hyoungsun Choi, Mohamed Jismy Aashik Rasool +1
Conformal regression provides finite-sample marginal coverage, but it does not by itself determine how interval width should adapt across heterogeneous inputs. Existing locally ada…
How Prompts Move Language Model Behavior: Frames, Salience, and Construal as Semantic Control
Dongseok Kim, Hyoungsun Choi, Mohamed Jismy Aashik Rasool +1
Prompt engineering is widely used to shape large language model behavior, yet it is often treated as a practical heuristic rather than as a form of natural-language control. This p…
-Table: A Statistical Explanation for Global SHAP
Dongseok Kim, Hyoungsun Choi, Mohamed Jismy Aashik Rasool +1
Global SHAP explanations are typically presented as feature-importance rankings, which identify variables that matter to a black-box model but do not indicate whether their effects…
Punching Above Precision: Small Quantized Model Distillation with Learnable Regularizer
Abdur Rehman, S M A Sharif, Md Abdur Rahaman +3
Quantization-aware training (QAT) combined with knowledge distillation (KD) is a promising strategy for compressing Artificial Intelligence (AI) models for deployment on resource-c…