BAAI/bge-m3 lost accuracy under flash attention until the Sentence Transformers patch
tomaarsen · x · 2026-07-23
BAAI/bge-m3 got noticeably worse under flash attention before the fix:
- STSB test Spearman: 0.8485 → 0.7239
- NanoBEIR mean nDCG@10: 0.6041 → 0.5414
After the patch, quality returns exactly while keeping the packing speedup.
Related event: Flash Attention Causes Performance Drops in BGE-M3 and Other Models(2 posts)→
More from Research
- Stanford HAI says Evo 2 underpins the largest open language model and AI-made genomes — StanfordHAI · 2026-07-23
- MagNET predicts NMR in seconds as an ML surrogate for DFT calculations — CatAstro_Piyush · 2026-07-23
- Domenic Denicola shares his July 2026 agentic coding setup — cnakazawa · 2026-07-23
- A new notebook shows learning-rate boundaries can be fractal — S_Conradi · 2026-07-23
- Production agentic systems need a context layer, not just LLM tool calls — Pavan_Belagatti · 2026-07-23
- Nature asks whether organoid intelligence could become the next computing paradigm — LadiesMan-8 · 2026-07-23