ancient dna reconstruction 1corruption-aware training 1genomic transformers 1knowledge-guided masking 1masked language modeling 1
From the 1 of 2 linked papers with an AI index.
2 papers
cs.LG2026
VESTIGE: A Knowledge-Guided Masking Strategy for Corruption-Aware Fine-Tuning of Genomic Transformers, Validated on Ancient DNA Reconstruction
Angshuman Chakravertty, Rahul Maheshwari
The paper presents VESTIGE, a parameter‑free masking strategy that aligns masked language model training with measured per‑position DNA damage profiles, improving ancient DNA recon…
cs.AR2026
Benchmarking LLMs for Verilog Design Flows
Angshuman Chakravertty, Rahul Koshti, Buddhi Prakash Sharma +1
Large language models (LLMs) show promise in code generation, but their capabilities to produce correct, synthesizable hardware description language (HDL) code still remain to be p…