Results for “speculative-decoding”

Reduce per-token latency on Jetson vLLM servers by appending speculative decoding configuration, with guidance on when to enable and how to benchmark the improvement.

Evaluates image captions by converting them into scene graphs and computing an F-score over semantic propositions, measuring how well a generated caption captures the meaning of an image compared to human references.

Validates the Emotional Attitude Score (EAS) metric by measuring its consistency with human judgment on word-level sentiment polarity, using the AmbGIMT dataset and pairwise score comparisons.

Evaluates text-to-image generation models by decomposing assessment into five specialized skills (object presence, count, spatial relations, scale, and text rendering) and open-ended prompts, producing interpretable binary scores with visual and textual explanations.

Evaluates how well internal model representations (hidden states) predict token-level information importance in summarization tasks, using NDCG@10 and Spearman's rank correlation.

Explains machine learning model predictions using SHAP values, covering feature importance, visualization plots, model debugging, bias analysis, and production deployment.

Explains machine learning model predictions using SHAP values, covering feature importance, visualizations, debugging, bias analysis, and production deployment.

Train large-scale MoE models with FP8/INT4 low-precision RL, speculative decoding, and train-inference alignment using the miles framework.