idiap/sdialog

RAFT, or Retrieval-Augmented Fine-Tuning, is a method comprising of a fine-tuning and a RAG-based retrieval phase. It is particularly suited for the creation of agents that realistically emulate a specific human target.

An easy to use off-policy synthetic data distillation engine for LLMs.

DialogLab is an authoring tool for configuring and running Human-AI multi-party conversations.

A framework for training and evaluating AI models on a variety of openly available dialogue datasets.

NeMo Guardrails is an open-source toolkit for easily adding programmable guardrails to LLM-based conversational systems.

DataDreamer: Prompt. Generate Synthetic Data. Train & Align Models.   🤖💤

EMNLP 2026 Main Conference Paper: MASCOT: Towards Multi-Agent Socio-Collaborative Companion Systems (https://aclanthology.org/2026.emnlp-main.878/))

A TTS model capable of generating ultra-realistic dialogue in one pass.

A project to improve skills of large language models