1.3k citations · 1.8k across the 110 of their papers we have counts for
26 papers · 1 filter
Molt: A Scalable PyTorch-Native Training Framework for Agentic Reinforcement Learning
Jian Hu, Huiying Li, Hao Zhang +8
Agentic reinforcement learning requires rapid experimentation with agents and learning algorithms, yet large policies and long, multimodal trajectories demand substantial distribut…
X-Token: Projection-Guided Cross-Tokenizer Knowledge Distillation
Sharath Turuvekere Sreenivas, Adithyakrishna Venkatesh Hanasoge, Mingyu Yang +4
Cross-tokenizer knowledge distillation allows a student model to learn from teachers with incompatible vocabularies. Prior work operates on hidden states or logits; the latter is p…
Nemotron 3 Nano Omni: Efficient and Open Multimodal Intelligence
NVIDIA, :, Amala Sanjay Deshmukh +204
We introduce Nemotron 3 Nano Omni, the latest model in the Nemotron multimodal series and the first to natively support audio inputs alongside text, images, and video. Nemotron 3 N…
Star Elastic: Many-in-One Reasoning LLMs with Efficient Budget Control
Ali Taghibakhshi, Ruisi Cai, Saurav Muralidharan +17
Training a family of large language models (LLMs), either from scratch or via iterative compression, is prohibitively expensive and inefficient, requiring separate training runs fo…
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…
GSPN-2: Efficient Parallel Sequence Modeling
Hongjun Wang, Yitong Jiang, Collin McCarthy +12
Efficient vision transformer remains a bottleneck for high-resolution images and long-video related real-world applications. Generalized Spatial Propagation Network (GSPN) addresse…
Nemotron-Flash: Towards Latency-Optimal Hybrid Small Language Models
Yonggan Fu, Xin Dong, Shizhe Diao +12
Efficient deployment of small language models (SLMs) is essential for numerous real-world applications with stringent latency constraints. While previous work on SLM design has pri…
NVIDIA Nemotron Nano V2 VL
NVIDIA, :, Amala Sanjay Deshmukh +121
We introduce Nemotron Nano V2 VL, the latest model of the Nemotron vision-language series designed for strong real-world document understanding, long video comprehension, and reaso…
DLER: Doing Length pEnalty Right - Incentivizing More Intelligence per Token via Reinforcement Learning
Shih-Yang Liu, Xin Dong, Ximing Lu +9
Reasoning language models such as OpenAI-o1, DeepSeek-R1, and Qwen achieve strong performance via extended chains of thought but often generate unnecessarily long outputs. Maximizi…
BroRL: Scaling Reinforcement Learning via Broadened Exploration
Jian Hu, Mingjie Liu, Ximing Lu +8
Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as a key ingredient for unlocking complex reasoning capabilities in large language models. Recent work ProRL has s…
LaCache: Ladder-Shaped KV Caching for Efficient Long-Context Modeling of Large Language Models
Dachuan Shi, Yonggan Fu, Xiangchi Yuan +8
Recent advancements in Large Language Models (LLMs) have spurred interest in numerous applications requiring robust long-range capabilities, essential for processing extensive inpu…
Minifinetuning: Low-Data Generation Domain Adaptation through Corrective Self-Distillation
Peter Belcak, Greg Heinrich, Jan Kautz +1
Finetuning language models for a new domain inevitably leads to the deterioration of their general performance. This becomes more pronounced the more limited the finetuning data re…
TwinTURBO: Semi-Supervised Fine-Tuning of Foundation Models via Mutual Information Decompositions for Downstream Task and Latent Spaces
Guillaume Quétant, Pavlo Molchanov, Slava Voloshynovskiy
We present a semi-supervised fine-tuning framework for foundation models that utilises mutual information decomposition to address the challenges of training for a limited amount o…
Advancing Weight and Channel Sparsification with Enhanced Saliency
Xinglong Sun, Maying Shen, Hongxu Yin +3
Pruning aims to accelerate and compress models by removing redundant parameters, identified by specifically designed importance scores which are usually imperfect. This removal is…
Entropy-Regularized Process Reward Model
Hanning Zhang, Pengcheng Wang, Shizhe Diao +6
Large language models (LLMs) have shown promise in performing complex multi-step reasoning, yet they continue to struggle with mathematical reasoning, often making systematic error…
Puzzle: Distillation-Based NAS for Inference-Optimized LLMs
Akhiad Bercovich, Tomer Ronen, Talor Abramovich +23
Large language models (LLMs) offer remarkable capabilities, yet their high inference costs restrict wider adoption. While increasing parameter counts improves accuracy, it also bro…
PHI-S: Distribution Balancing for Label-Free Multi-Teacher Distillation
Mike Ranzinger, Jon Barker, Greg Heinrich +3
Various visual foundation models have distinct strengths and weaknesses, both of which can be improved through heterogeneous multi-teacher knowledge distillation without labels, te…
A deeper look at depth pruning of LLMs
Shoaib Ahmed Siddiqui, Xin Dong, Greg Heinrich +4
Large Language Models (LLMs) are not only resource-intensive to train but even more costly to deploy in production. Therefore, recent work has attempted to prune blocks of LLMs bas…
Adaptive Sharpness-Aware Pruning for Robust Sparse Networks
Anna Bair, Hongxu Yin, Maying Shen +2
Robustness and compactness are two essential attributes of deep learning models that are deployed in the real world. The goals of robustness and compactness may seem to be at odds,…
Do Gradient Inversion Attacks Make Federated Learning Unsafe?
Ali Hatamizadeh, Hongxu Yin, Pavlo Molchanov +8
Federated learning (FL) allows the collaborative training of AI models without needing to share raw data. This capability makes it especially interesting for healthcare application…
Privacy Vulnerability of Split Computing to Data-Free Model Inversion Attacks
Xin Dong, Hongxu Yin, Jose M. Alvarez +3
Mobile edge devices see increased demands in deep neural networks (DNNs) inference while suffering from stringent constraints in computing resources. Split computing (SC) emerges a…
See through Gradients: Image Batch Recovery via GradInversion
Hongxu Yin, Arun Mallya, Arash Vahdat +3
Training deep neural networks requires gradient estimation from data batches to update parameters. Gradients per parameter are averaged over a set of data and this has been presume…
HarDNN: Feature Map Vulnerability Evaluation in CNNs
Abdulrahman Mahmoud, Siva Kumar Sastry Hari, Christopher W. Fletcher +7
As Convolutional Neural Networks (CNNs) are increasingly being employed in safety-critical applications, it is important that they behave reliably in the face of hardware errors. T…
Dreaming to Distill: Data-free Knowledge Transfer via DeepInversion
Hongxu Yin, Pavlo Molchanov, Zhizhong Li +5
We introduce DeepInversion, a new method for synthesizing images from the image distribution used to train a deep neural network. We 'invert' a trained network (teacher) to synthes…