7 citations · 17 across the 17 of their papers we have counts for
20 papers
Uncertainty principles as forcing mechanisms
Laurent Massoulié
In this note we first consider the Heisenberg uncertainty principle. We show that it acts as a {\bf forcing mechanism} in the following sense. Real functions that approach its limi…
Priming Base Models for Agentic Post-Training: An Open Midtraining Recipe
Soumye Singhal, Joseph Jennings, Ka Shun Shum +11
Frontier models, general-purpose and domain-specific alike, now include a midtraining stage between pretraining and post-training, but little has been published about how this stag…
Nemotron 3 Ultra: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning
NVIDIA, :, Aaron Blakeman +572
We introduce Nemotron 3 Ultra, a 550 billion total and 55 billion active parameter Mixture-of-Experts Hybrid Mamba-Attention language model. We pre-trained Nemotron 3 Ultra on 20 t…
Nemotron 3 Super: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning
NVIDIA, :, Aakshita Chandiramani +544
We describe the pre-training, post-training, and quantization of Nemotron 3 Super, a 120 billion (active 12 billion) parameter hybrid Mamba-Attention Mixture-of-Experts model. Nemo…
NVIDIA Nemotron 3: Efficient and Open Intelligence
NVIDIA, :, Aaron Blakeman +356
We introduce the Nemotron 3 family of models - Nano, Super, and Ultra. These models deliver strong agentic, reasoning, and conversational capabilities. The Nemotron 3 family uses a…
Nemotron 3 Nano: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning
NVIDIA, :, Aaron Blakeman +311
We present Nemotron 3 Nano 30B-A3B, a Mixture-of-Experts hybrid Mamba-Transformer language model. Nemotron 3 Nano was pretrained on 25 trillion text tokens, including more than 3 t…
Think Twice: Branch-and-Rethink Reasoning Reward Model
Yizhu Jiao, Jiaqi Zeng, Julien Veron Vialard +3
Large language models (LLMs) increasingly rely on thinking models that externalize intermediate steps and allocate extra test-time compute, with think-twice strategies showing that…
RLBFF: Binary Flexible Feedback to bridge between Human Feedback & Verifiable Rewards
Zhilin Wang, Jiaqi Zeng, Olivier Delalleau +6
Reinforcement Learning with Human Feedback (RLHF) and Reinforcement Learning with Verifiable Rewards (RLVR) are the main RL paradigms used in LLM post-training, each offering disti…
HelpSteer3-Preference: Open Human-Annotated Preference Data across Diverse Tasks and Languages
Zhilin Wang, Jiaqi Zeng, Olivier Delalleau +6
Preference datasets are essential for training general-domain, instruction-following language models with Reinforcement Learning from Human Feedback (RLHF). Each subsequent data re…
Llama-Nemotron: Efficient Reasoning Models
Akhiad Bercovich, Itay Levy, Izik Golan +132
We introduce the Llama-Nemotron series of models, an open family of heterogeneous reasoning models that deliver exceptional reasoning capabilities, inference efficiency, and an ope…
Adversarial Training of Reward Models
Alexander Bukharin, Haifeng Qian, Shengyang Sun +6
Reward modeling has emerged as a promising approach for the scalable alignment of language models. However, contemporary reward models (RMs) often lack robustness, awarding high re…
Nemotron-H: A Family of Accurate and Efficient Hybrid Mamba-Transformer Models
NVIDIA, :, Aaron Blakeman +198
As inference-time scaling becomes critical for enhanced reasoning capabilities, it is increasingly becoming important to build models that are efficient to infer. We introduce Nemo…
HelpSteer3: Human-Annotated Feedback and Edit Data to Empower Inference-Time Scaling in Open-Ended General-Domain Tasks
Zhilin Wang, Jiaqi Zeng, Olivier Delalleau +6
Inference-Time Scaling has been critical to the success of recent models such as OpenAI o1 and DeepSeek R1. However, many techniques used to train models for inference-time scaling…
Reward-aware Preference Optimization: A Unified Mathematical Framework for Model Alignment
Shengyang Sun, Yian Zhang, Alexander Bukharin +11
The rapid development of large language model (LLM) alignment algorithms has resulted in a complex and fragmented landscape, with limited clarity on the effectiveness of different…
Diverging Preferences: When do Annotators Disagree and do Models Know?
Michael JQ Zhang, Zhilin Wang, Jena D. Hwang +6
We examine diverging preferences in human-labeled preference datasets. We develop a taxonomy of disagreement sources spanning ten categories across four high-level classes and find…
HelpSteer2-Preference: Complementing Ratings with Preferences
Zhilin Wang, Alexander Bukharin, Olivier Delalleau +5
Reward models are critical for aligning models to follow instructions, and are typically trained following one of two popular paradigms: Bradley-Terry style or Regression style. Ho…
HelpSteer2: Open-source dataset for training top-performing reward models
Zhilin Wang, Yi Dong, Olivier Delalleau +6
High-quality preference datasets are essential for training reward models that can effectively guide large language models (LLMs) in generating high-quality responses aligned with…
Nemotron-4 340B Technical Report
Nvidia, :, Bo Adler +80
We release the Nemotron-4 340B model family, including Nemotron-4-340B-Base, Nemotron-4-340B-Instruct, and Nemotron-4-340B-Reward. Our models are open access under the NVIDIA Open…
NeMo-Aligner: Scalable Toolkit for Efficient Model Alignment
Gerald Shen, Zhilin Wang, Olivier Delalleau +10
Aligning Large Language Models (LLMs) with human values and preferences is essential for making them helpful and safe. However, building efficient tools to perform alignment can be…
HelpSteer: Multi-attribute Helpfulness Dataset for SteerLM
Zhilin Wang, Yi Dong, Jiaqi Zeng +8
Existing open-source helpfulness preference datasets do not specify what makes some responses more helpful and others less so. Models trained on these datasets can incidentally lea…