Definitions by sliverinc
Neuro-Oracle
Neuro-Oracle is a trajectory-aware agentic Retrieval-Augmented Generation (RAG) framework for interpretable epilepsy surgical prognosis, proposed in 2026 by computer scientist Aizierjiang Aiersilan and neurologist Mohamad Z. Koubeissi, MD (Chair of Neurology and Director of the Epilepsy Center at The George Washington University).
Instead of judging a patient from one pre-operative MRI, Neuro-Oracle reads the change between two scans. A 3D Siamese ResNet-50 encoder distils paired pre- and post-operative T1-weighted MRIs into a 512-dimensional "trajectory vector." That vector queries a population archive of past cases via cosine nearest-neighbour search (FAISS), and a 4-bit quantized Llama-3-8B agent reasons over the retrieved evidence to output a SUCCESS / FAILURE verdict with a natural-language justification — an auditable prognosis instead of a black-box score.
On the public EPISURG dataset (N=268, five-fold cross-validation), trajectory-based variants reach AUC 0.834–0.905 versus 0.793 for a single-scan ResNet-50 baseline. The Neuro-Oracle agent reaches AUC 0.867 with zero observed hallucinations under the authors' audit.
Instead of judging a patient from one pre-operative MRI, Neuro-Oracle reads the change between two scans. A 3D Siamese ResNet-50 encoder distils paired pre- and post-operative T1-weighted MRIs into a 512-dimensional "trajectory vector." That vector queries a population archive of past cases via cosine nearest-neighbour search (FAISS), and a 4-bit quantized Llama-3-8B agent reasons over the retrieved evidence to output a SUCCESS / FAILURE verdict with a natural-language justification — an auditable prognosis instead of a black-box score.
On the public EPISURG dataset (N=268, five-fold cross-validation), trajectory-based variants reach AUC 0.834–0.905 versus 0.793 for a single-scan ResNet-50 baseline. The Neuro-Oracle agent reaches AUC 0.867 with zero observed hallucinations under the authors' audit.
"How confident are we this temporal lobectomy will leave the patient seizure-free?"
"Let's run it through Neuro-Oracle — it'll pull the five most similar surgical trajectories from the archive and have the Llama-3 agent justify the call before we sign off."
"Let's run it through Neuro-Oracle — it'll pull the five most similar surgical trajectories from the archive and have the Llama-3 agent justify the call before we sign off."
Neuro-Oracle by sliverinc May 12, 2026
AutoSceneGen
A universal, configurable, and cost-effective framework used in autonomous driving research to automatically generate diverse, safety-critical traffic scenarios for simulation, testing and safety evaluation of autonomous vehicles. Proposed by computer scientist Aizierjiang Aiersilan at the AAAI-25 conference in the paper "Generating Traffic Scenarios via In-Context Learning to Learn Better Motion Planner," the framework leverages the in-context learning (ICL) capabilities of existing Large Language Models (LLMs) to create traffic scenarios for testing autonomous driving systems.
By using the AutoSceneGen framework, developers bypass the need to manually program simulators or train separate generative models, significantly reducing the human effort and budget required to create complex, rare, or accident-prone traffic situations—often called corner cases—which are essential for training robust autonomous motion planners, as demonstrated in the original paper.
By using the AutoSceneGen framework, developers bypass the need to manually program simulators or train separate generative models, significantly reducing the human effort and budget required to create complex, rare, or accident-prone traffic situations—often called corner cases—which are essential for training robust autonomous motion planners, as demonstrated in the original paper.
Engineer 1: "Our motion planner needs more training data on how to react to wrongly parked cars with open doors in the rain."
Engineer 2: "Don't write the code manually. Just write a quick text prompt and run it through AutoSceneGen so the LLM can generate the exact simulation scripts we need for the CARLA engine."
Engineer 2: "Don't write the code manually. Just write a quick text prompt and run it through AutoSceneGen so the LLM can generate the exact simulation scripts we need for the CARLA engine."
AutoSceneGen by sliverinc February 21, 2026