openbmb/MiniCPM5-1B
Text Generation • 1B • Updated • 2 • 217
Keylhan Paumard--André
keypa
AI & ML interests
Efficient deep learning, LLM fine-tuning, inference optimization, model compression, distributed training, GPU systems, open-source AI infrastructure
Recent Activity
liked a model about 9 hours ago
openbmb/MiniCPM5-1B liked a model about 18 hours ago
sakamakismile/Huihui-Qwen3.6-27B-abliterated-NVFP4-MTP liked a model about 18 hours ago
sakamakismile/Qwen3.6-27B-Text-NVFP4-MTPOrganizations
liked a
model about 9 hours ago
liked 3
models about 18 hours ago
liked a
model 2 days ago
liked a
dataset 2 days ago
upvoted a collection 2 days ago
liked a
model 2 days ago
liked a
model 3 days ago
liked 2
datasets 4 days ago
liked a
model 4 days ago
liked a
dataset 4 days ago
liked a
dataset 5 days ago
liked a
model 6 days ago
upvoted 2
articles 6 days ago
Article
Introducing the Ettin Reranker Family
tomaarsen
• • 45 Article
From Zero to GPU: A Guide to Building and Scaling Production-Ready CUDA Kernels
drbh, danieldk
• • 100 reacted to salma-remyx's post with 🚀 6 days ago
Post
5083
Just trained a 2B coding model to rank candidate AI/ML research ideas against the implicit preferences in a code repository's merge history.
The training data comes from a Gaussian Process fit on the accumulated dispositions in VQASynth, where each PR against a deployed project yields a pairwise comparison between the feature branch preferred and the baseline at main.
The GP scores candidate papers to synthesize preference pairs, and DPO with LoRA bakes the ranking pipeline into the model's weights.
After 1 epoch the model reaches 87.4% reward accuracy on the held-out eval split against 92.3% on training, consistent with learning the task without overfitting.
Now, I'm scaling the pipeline to thousands of repos for a generalization test.
Dataset: remyxai/mhpd-dpo-v0
Model: remyxai/mhpd-dpo-qwen3.5-2b-vqasynth
Substack: https://remyxai.substack.com/p/the-ai-pm
The training data comes from a Gaussian Process fit on the accumulated dispositions in VQASynth, where each PR against a deployed project yields a pairwise comparison between the feature branch preferred and the baseline at main.
The GP scores candidate papers to synthesize preference pairs, and DPO with LoRA bakes the ranking pipeline into the model's weights.
After 1 epoch the model reaches 87.4% reward accuracy on the held-out eval split against 92.3% on training, consistent with learning the task without overfitting.
Now, I'm scaling the pipeline to thousands of repos for a generalization test.
Dataset: remyxai/mhpd-dpo-v0
Model: remyxai/mhpd-dpo-qwen3.5-2b-vqasynth
Substack: https://remyxai.substack.com/p/the-ai-pm
liked a
model 6 days ago
liked a
model 7 days ago