Suyash Shringarpure (@suyashss) profile picture

Scientist at 23andMe. Interests: Machine learning, Genetics, Privacy

Joined September 2008
Photo by Suyash Shringarpure (@suyashss)Photo by Suyash Shringarpure (@suyashss)
Suyash Shringarpure retweeted
Introducing GPT-Rosalind, our frontier reasoning model built to support research across biology, drug discovery, and translational medicine.
476
1,246
450
12,642
2,357,639
Suyash Shringarpure retweeted
Delighted to share our latest research from the @23andMeResearch Team, just published in @Nature ! We looked at data from >27,000 participants to uncover how human genetics influences weight loss efficacy and side effects of GLP-1 medications like semaglutide. A thread 🧵👇
14
66
10
236
60,789
🚨New work led by #CZBiohubCHI Investigator Aly Khan & @23andMeResearch accepted to #NeurIPS2025. PRSformer: disease risk prediction from genetic variants at population scale! New scaling law (N>1M) + multitask + multi-ancestry gains for 18 immune traits. biorxiv.org/content/10.1101/…
2
11
1
21
5,531
imho, GSM8k is not a proper RL demo task for these papers. i propose the Sudoku task that @HrishbhDalal wrote a GRPO blogpost about. it's not the same style of data that the Qwen team annealed on, and it only works at ~7b+ scale (this applies to most complex RL tasks for LLMs)
GRPO just got a speed boost! Xiamen University introduced Completion Pruning Policy Optimization (CPPO), which significantly reduces the number of gradient calculations and updates. How fast? On GSM8K, it's 8.32× faster than GRPO, and on MATH, the speedup is 3.51×. 🚀🔥
7
2
1
116
14,776
Similar experience trying gene ranking with RL like in the OpenAI demo. I found that 8b models work well, but smaller models need multiple rewards that can be hacked, and don’t do so well. suyashss.github.io/llm/2025/…
71
🧬 Just published a new blog post on reinforcement fine-tuning (RFT) for causal gene identification using Llama models! We show how a fine-tuned 8B parameter model can match the performance of a 70B model on this task. 🧵 (1/8)
1
1
6
1,040
Typo tagging @willcb , thanks for the simple GRPO notebook.
1
125
More tagging typos: Thanks to @charles_irl for the @modal_labs credits, and @HamelHusain and @dan_s_becker for their LLM fine-tuning course.
1
121
I built a tiny vision-language model using the MNIST digits dataset that can: ✅ Classify images → numbers ✅ Generate images from digit labels It combines image tokenization & nanoGPT for training.
1
3
185
This implementation is laptop-friendly, so you can run it on your laptop in a couple of hours even if you don't have access to specialized GPUs.
64
Suyash Shringarpure retweeted
The genetic architecture of gene expression in individuals of African and European ancestry medrxiv.org/cgi/content/shor… #medRxiv
2
2
488