Nemo guardrails: A toolkit for controllable and safe llm applications with programmable rails

T Rebedea, R Dinu, M Sreedhar, C Parisien… - arxiv preprint arxiv …, 2023 - arxiv.org
NeMo Guardrails is an open-source toolkit for easily adding programmable guardrails to
LLM-based conversational systems. Guardrails (or rails for short) are a specific way of …

Towards human-centered proactive conversational agents

Y Deng, L Liao, Z Zheng, GH Yang… - Proceedings of the 47th …, 2024 - dl.acm.org
Recent research on proactive conversational agents (PCAs) mainly focuses on improving
the system's capabilities in anticipating and planning action sequences to accomplish tasks …

Analyzing and Adapting Large Language Models for Few-Shot Multilingual NLU: Are We There Yet?

E Razumovskaia, I Vulić, A Korhonen - arxiv preprint arxiv:2403.01929, 2024 - arxiv.org
Supervised fine-tuning (SFT), supervised instruction tuning (SIT) and in-context learning
(ICL) are three alternative, de facto standard approaches to few-shot learning. ICL has …

Generative ai for synthetic data generation: Methods, challenges and the future

X Guo, Y Chen - arxiv preprint arxiv:2403.04190, 2024 - arxiv.org
The recent surge in research focused on generating synthetic data from large language
models (LLMs), especially for scenarios with limited data availability, marks a notable shift in …

Towards responsible development of generative AI for education: An evaluation-driven approach

I Jurenka, M Kunesch, KR McKee, D Gillick… - arxiv preprint arxiv …, 2024 - arxiv.org
A major challenge facing the world is the provision of equitable and universal access to
quality education. Recent advances in generative AI (gen AI) have created excitement about …

Mathdial: A dialogue tutoring dataset with rich pedagogical properties grounded in math reasoning problems

J Macina, N Daheim, SP Chowdhury, T Sinha… - arxiv preprint arxiv …, 2023 - arxiv.org
While automatic dialogue tutors hold great potential in making education personalized and
more accessible, research on such systems has been hampered by a lack of sufficiently …

Intent detection for task‐oriented conversational agents: A comparative study of recurrent neural networks and transformer models

M Jbene, A Chehri, R Saadane, S Tigani… - Expert …, 2025 - Wiley Online Library
Abstract Conversational assistants (CAs) and Task‐oriented ones, in particular, are
designed to interact with users in a natural language manner, assisting them in completing …

Multi 3 WOZ: A Multilingual, Multi-Domain, Multi-Parallel Dataset for Training and Evaluating Culturally Adapted Task-Oriented Dialog Systems

S Hu, H Zhou, M Hergul, M Gritta, G Zhang… - Transactions of the …, 2023 - direct.mit.edu
Creating high-quality annotated data for task-oriented dialog (ToD) is known to be
notoriously difficult, and the challenges are amplified when the goal is to create equitable …

Dialogstudio: Towards richest and most diverse unified dataset collection for conversational ai

J Zhang, K Qian, Z Liu, S Heinecke, R Meng… - arxiv preprint arxiv …, 2023 - arxiv.org
Despite advancements in conversational AI, language models encounter challenges to
handle diverse conversational tasks, and existing dialogue dataset collections often lack …

Mixture of soft prompts for controllable data generation

D Chen, C Lee, Y Lu, D Rosati, Z Yu - arxiv preprint arxiv:2303.01580, 2023 - arxiv.org
Large language models (LLMs) effectively generate fluent text when the target output follows
natural language patterns. However, structured prediction tasks confine the output format to …