Yanshuo Chen
@ys-chen.bsky.social
Computational biology Ph.D student @UMD. Interested in Generative Modeling.
🚨 Our paper got into ISMB/ECCB 2025!
We propose W₁OT — a fast, scalable solver for single-cell perturbation prediction 🚀
✅ 25–45× faster than CellOT
✅ Outperforms CPA, Biolord, CellOT
✅ Robust on unseen cell types 🧬📈
#ISMB2025 #scRNAseq #PerturbationPrediction
Paper: arxiv.org/abs/2411.00614
We propose W₁OT — a fast, scalable solver for single-cell perturbation prediction 🚀
✅ 25–45× faster than CellOT
✅ Outperforms CPA, Biolord, CellOT
✅ Robust on unseen cell types 🧬📈
#ISMB2025 #scRNAseq #PerturbationPrediction
Paper: arxiv.org/abs/2411.00614
April 21, 2025 at 4:58 PM
🚨 Our paper got into ISMB/ECCB 2025!
We propose W₁OT — a fast, scalable solver for single-cell perturbation prediction 🚀
✅ 25–45× faster than CellOT
✅ Outperforms CPA, Biolord, CellOT
✅ Robust on unseen cell types 🧬📈
#ISMB2025 #scRNAseq #PerturbationPrediction
Paper: arxiv.org/abs/2411.00614
We propose W₁OT — a fast, scalable solver for single-cell perturbation prediction 🚀
✅ 25–45× faster than CellOT
✅ Outperforms CPA, Biolord, CellOT
✅ Robust on unseen cell types 🧬📈
#ISMB2025 #scRNAseq #PerturbationPrediction
Paper: arxiv.org/abs/2411.00614
Reposted by Yanshuo Chen
Watermarking LLMs is crucial for traceability, and speculative sampling accelerates generation.
Can we have the best of both worlds?🤔
Our research shows it's not possible😮—we prove a no-go theorem and propose methods to navigate this trade-off. #NeurIPS2024
nips.cc/virtual/2024...
Can we have the best of both worlds?🤔
Our research shows it's not possible😮—we prove a no-go theorem and propose methods to navigate this trade-off. #NeurIPS2024
nips.cc/virtual/2024...
December 12, 2024 at 6:13 AM
Watermarking LLMs is crucial for traceability, and speculative sampling accelerates generation.
Can we have the best of both worlds?🤔
Our research shows it's not possible😮—we prove a no-go theorem and propose methods to navigate this trade-off. #NeurIPS2024
nips.cc/virtual/2024...
Can we have the best of both worlds?🤔
Our research shows it's not possible😮—we prove a no-go theorem and propose methods to navigate this trade-off. #NeurIPS2024
nips.cc/virtual/2024...