Beomseok Kang
- Papers
- 4
Cite
Notes
Only stored in your browser.
4papers
Authored papers
4CompactAttention: Accelerating Chunked Prefill with Block-Union KV Selection
arXiv 2026
Token Sparse Attention: Efficient Long-Context Inference with Interleaved Token Selection
arXiv 2026
QWHA: Quantization-Aware Walsh-Hadamard Adaptation for Parameter-Efficient Fine-Tuning on Large Language Models
arXiv 2025
LiteStage: Latency-aware Layer Skipping for Multi-stage Reasoning
arXiv 2025
Affiliations
No known affiliations.
Frequent co-authors
6from 4 papers