HEED replaces uniform residual alignment with density-weighted alignment using patch self-dissimilarity to improve hybrid VLM distillation, gaining 8.7 points on OCRBench v2 and 5.13 on a 10-benchmark average.
Distilling the knowledge in a neural network
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
BDTS is a new data-structural framework for budgeted maintenance of rooted trace graphs, with Rust benchmarks showing compaction of 350k-2.71M tokens to 1k-4k tokens and model input reduction from ~3360 to ~432 tokens.
citing papers explorer
-
HEED: Density-Weighted Residual Alignment for Hybrid Vision-Language Model Distillation
HEED replaces uniform residual alignment with density-weighted alignment using patch self-dissimilarity to improve hybrid VLM distillation, gaining 8.7 points on OCRBench v2 and 5.13 on a 10-benchmark average.
-
Budgeted Dynamic Trace Structures for Token-Efficient Sequential Computation
BDTS is a new data-structural framework for budgeted maintenance of rooted trace graphs, with Rust benchmarks showing compaction of 350k-2.71M tokens to 1k-4k tokens and model input reduction from ~3360 to ~432 tokens.