Stanford's Kundaje publicly disputes TraceBind paper for not citing or comparing ChromBPNet
NancyZh60672287 · x · 2026-09-23
Stanford researcher Anshul Kundaje publicly questioned why the TraceBind paper on Tn5 bias correction of ATAC-seq neither cites nor compares against ChromBPNet.
The authors responded that the two do different things: ChromBPNet is a sequence-to-accessibility prediction model, while TraceBind takes ATAC data for footprint analysis, with different inputs, outputs, and goals. Kundaje pushed back that anyone who has read the ChromBPNet paper would know that claim doesn't hold.
A notable open dispute over method credit in genomics ML.
More from Research
- Trust AI R&D evals only if scored by people who've hand-labeled outputs — dfrsrchtwts · 2026-09-23
- JevBench Model Decision Index Debuts on Hugging Face, Native Image Support Still Missing — openSourcerer9000 · 2026-09-23
- AI swarms spontaneously grow scale-free hub topologies, MIT professor observes — ProfBuehlerMIT · 2026-09-23
- Flex-π: a 6B world-action model beats π0.5 by up to 6x on real bimanual tasks — chris_j_paxton · 2026-09-23
- Deoptimization Is Harder Than Optimization: How JIT Escape Hatches Keep Dynamic Languages Fast — lauriewired · 2026-09-23
- Up to 58% of DFT-generated data is physically impossible, Valence Labs warns — josephdviviano · 2026-09-23