-
Neuro-symbolic AI is a type of [artificial intelligence](https://en.wikipedia.org/wiki/Artificial_intelligence) that integrates [neural](https://en.wikipedia.org/wiki/Connectionism) and [symbolic AI](…
-
I took the [Phi 4 GRPO](https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Phi_4_(14B)-GRPO.ipynb) notebook and switched out the model for `Phi 3 Mini 128k Instruct`, had to dis…
-
Hi, thanks for a very cool collection!
These may be relevant, for Neuro Symbolic AI
https://lis.csail.mit.edu/bilevel-planning-for-robots-an-illustrated-introduction
https://github.com/Learning-and-In…
-
This Issue explores how scalar phase drift can serve as a framework for improving AI reasoning, memory, and test-time inference, inspired by recent work on System 2 thinking.
📘 Context:
Lilian Weng (…
-
There is a trust zone in the example threat model :
```
{
"symbolic_name": "public",
"title": "Public Internet",
"description": "The public internet"
…
-
triton-windows-3.3.0a0.post17
2.7.0+cu128
python 3.12.9
So since i have updated sage etc.. i can`t make them work together, instead torch compile alone does just work but not together with block cac…
-
A few of the controls in the HuskyAI example reference missing trust zones, e.g.
https://github.com/OWASP/www-project-threat-model-library/blob/main/threat-models/husky-ai-threat-model.json#L931-L934…
-
Seems " this module "torch.onnx.symbolic_registry" has been replaced and Vitis ai copyleft for yolov6 quantization is no longer working.
-
### 🐛 Describe the bug
```python
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
from torch_mlir import fx
model_name = "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B"
tokenize…
-
Hi AMAI Team,
I wanted to start by commending you on the great work you have produced. However, it appears m to me that certain aspects and technologies crucial to important AI subfields like Symboli…