Unverified50% confidenceBenchmarkExact time
In protein-ligand cofolding experiments, using the Boltz flow map version with binding affinity as the oracle reward, the success criterion (RMSD less than 2 angstroms and PB valid above threshold) improved from 61.8% to 64.5%.
1
Sources
50%
Confidence
Long-term
Relevance
11/2/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedAlphaFold2 solved the protein structure prediction problem in 2020, achieving experimental-level prediction accuracy at the CASP14 competition with a median GDT score exceeding 92.65% similarUnverifiedAlphaFold2 combines multiple sequence alignment (MSA) with equivariant neural networks, and for the first time achieved experimental-level prediction accuracy on the majority of proteins, with a median TM-score exceeding 0.9.65% similarUnverifiedIn two-dimensional and eight-dimensional Gaussian mixture experiments, the method shows performance superior to traditional score-matching generators—at similar parameter counts, the number of off-mode samples is significantly reduced.62% similarUnverified模型量化(将权重从FP16压缩为INT8或INT4)虽然在大多数基准测试上性能损失很小,但某些特定token生成概率可能发生显著变化,导致边缘情况下产生不同输出62% similarUnverifiedExperimental results show that the transferred performance of DP-μP matches or even exceeds per-model individually tuned baselines while significantly reducing computational cost.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1127431API
curl https://kongchang.com/api/v1/knowledge/claims/1127431MCP
get_claim(id=1127431)