Orchestra-Research

Guarantee valid JSON/XML/code structure during generation, use Pydantic models for type-safe outputs, support local models (Transformers…

Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods.

Meta's 86M prompt injection and jailbreak detector.

High-level PyTorch framework with Trainer class, automatic distributed training (DDP/FSDP/DeepSpeed), callbacks system, and minimal…

Provides guidance for performing causal interventions on PyTorch models using pyvene's declarative intervention framework.

High-performance vector similarity search engine for RAG and semantic search.

Scalable data processing for ML workloads.

Distributed training orchestration across clusters.

Guides researchers through structured ideation frameworks to discover high-impact research directions.

RNN+Transformer hybrid with O(n) inference.

Provides guidance for training and analyzing Sparse Autoencoders (SAEs) using SAELens to decompose neural network activations into…

Foundation model for image segmentation with zero-shot transfer.