Artificial Intelligence thread
- Thread starter 9dashline
- Start date
sure and my response to that is that they already have enough AI chips for next 2 years. And >5 million per year is just purely wasteful at this point. This seems more ignorance on his part.
of course, why do people keep asking me these questions.
I said they have enough.
exactly, training is clearly not a problem. So inference is needed and the Ascend chips are fine for inference.
OpenBMB releases multi-modal model MiniCPM-o 2.6 for edge devices like iPad
Also available on huggingface and github
comparable to gpt-4o-202405
my piece on the recent AI chip restrictions imposed by US government. This restricted access and closed system is bad for competitiveness in global market.
DeepSeek R1 (non-lite version) is essentially onpar with o1
Last edited by a moderator:
GDiffRetro: Retrosynthesis Prediction with Dual Graph Enhanced Molecular Representation and Diffusion Generation.
Abstract
Retrosynthesis prediction focuses on identifying reactants capable of synthesizing a target product. Typically, the retrosynthesis prediction involves two phases: Reaction Center Identification and Reactant Generation. However, we argue that most existing methods suffer from two limitations in the two phases: (i) Existing models do not adequately capture the “face” information in molecular graphs for the reaction center identification. (ii) Current approaches for the reactant generation predominantly use sequence generation in a 2D space, which lacks versatility in generating reasonable distributions for completed reactive groups and overlooks molecules’ inherent 3D properties. To overcome the above limitations, we propose GDiffRetro. For the reaction center identification, GDiffRetro uniquely integrates the original graph with its corresponding dual graph to represent molecular structures, which helps guide the model to focus more on the faces in the graph. For the reactant generation, GDiffRetro employs a conditional diffusion model in 3D to further transform the obtained synthon into a complete reactant. Our experimental findings reveal that GDiffRetro outperforms state-of-the-art semi-template models across various evaluative metrics.
