AVI Variant Impact Score Matches SOTA on Benchmarks, Though Evals Are Biased
anshulkundaje · x · 2026-09-18
In a long thread, Stanford professor Anshul Kundaje assesses the AVI (AlphaGenome Variant Impact) score from the AlphaGenome team. He notes AVI is very competitive with SOTA variant prioritization scores across established rare- and common-variant benchmarks — marginally better in some cases, substantially better in others. He cautions that these widely used evals carry biases and don't fully reveal each method's strengths and limits, but says AVI is not fundamentally flawed.
More from Research
- Idea: replicate JEV-style probability estimates via softmax of LLM logprobs — ngxson · 2026-09-18
- msign Is Wrong for Full-Rank SVD Power Iteration, but Correct for Top-k Subspaces — YouJiacheng · 2026-09-18
- Human Organoids Fill 91.9% of Cortex in Engineered Mice, Nature Paper Reports — ctjlewis · 2026-09-18
- Lambda's EdiVal-Agent uses an agentic judge for image editing evals, hitting 81.3% human agreement — TheZachMueller · 2026-09-18
- Tahoe-100M cell state dataset published in Cell, downloaded 600K+ times — anshulkundaje · 2026-09-18
- AlphaGenome author pushes back: variant prioritization gains are real contributions — anshulkundaje · 2026-09-18