Papersby GlyphNVIDIA research papers
A company publishing mostly on multimodal and robotics, best known here for Megatron-LM.90 papers · 6,925 citations · latest Nov 2026
VFlash: Fast, Accurate Visual Planning via Flash Primitive DecodingA method called VFlash speeds up and improves visual motion planning for robots by predicting entire spatial shapes at once.
NVIDIA109 upvotesRead →Nemotron-Labs-Diffusion-Uni: Elastic Self-Refining Unified Diffusion LLM with Cross-Paradigm On-Policy DistillationNLD-Uni is a unified discrete diffusion language model that combines visual understanding and generation via cross-paradigm on-policy distillation and progressive self-refining diffusion.NVIDIA12 upvotesRead →ThinkWorld: World Imagination Reinforces Chain-of-Dynamics ReasoningThinkWorld improves robotic generalization by representing actions as dynamics chains and using world imagination to learn from unlabeled human videos.NVIDIA57 upvotesRead →Joints, Camera, Action! Co-Generating 3D Motion and Video for Arbitrary RigsThis model generates both video and 3D motion for any character rig, allowing the motion to be replayed accurately within a 3D scene.NVIDIA26 upvotesRead →EdgeWise: Bringing Long-Horizon Agents to the Edge via Selective Cloud DelegationThis paper introduces a framework that reduces inference costs by using a small local model for most tasks and delegating only critical steps to the cloud.NVIDIA30 upvotesRead →When Do We Need On-Policy Distillation? Distilling on Offline Student Rollouts Is Often BetterThis paper shows that distilling from offline student rollouts is often more efficient and accurate than standard on-policy distillation for many model pairs.NVIDIA28 upvotesRead →Long-WAM: Scaling the Context of World-Action ModelsLong-WAM improves robot control by using autoregressive pretraining to effectively leverage long visual histories for real-time action prediction.NVIDIA2 citations114 upvotesRead →NeMo-DCR: Bit-Exact Delta-Compressed Refit for Scalable Agentic RL at Trillion-Parameter ScaleNeMo-DCR enables efficient, bit-exact synchronization of trillion-parameter model checkpoints by transmitting only the sparse weight changes between clusters.NVIDIA1 citations29 upvotesRead →UNREAL: Unifying Retrieval and Long-Context with a Single ModelUNREAL uses a single model-internal mechanism to perform both corpus retrieval and long-context evidence selection, improving performance while reducing computational costs.NVIDIA31 upvotesRead →LoGRA: Scaling LLM Reinforcement Learning with Low-Rank Gradient SketchesLoGRA reduces memory usage during reinforcement learning by using low-rank gradient sketches and step control to stabilize training.NVIDIA153 upvotesRead →Mid-Harness: Scaling Actions Between Model and Harness for Terminal AgentsMid-Harness improves agent reliability by sampling and verifying candidate actions before execution, rather than relying on a single generation.NVIDIA3 citations120 upvotesRead →LongLive-Plug: Once-for-All Distillation for Video GenerationLongLive-Plug distills reusable capabilities into LoRAs that can be deployed on diverse video generation models without retraining.NVIDIA1 citations42 upvotesRead →PixelUMM: Encoder-Free Unified Image and Video Understanding and GenerationPixelUMM is an encoder-free model that unifies image and video understanding and generation directly in pixel space.NVIDIA2 citations33 upvotesRead →An RL View of OPD: Least Square Policy Distillation for Sample-Efficient LLM ReasoningLSPD applies reinforcement learning principles to policy distillation, improving sample efficiency and policy diversity in language model training.NVIDIA26 upvotesRead →OSWorld-Pro: Process-based Evaluation for Computer Use AgentsOSWorld-Pro provides a procedural evaluation framework for computer-use agents by breaking tasks into subgoals and tracking progress.NVIDIA28 upvotesRead →SoL-Pi: Recursively Scaling Auto-Research Loops for Efficient Agent HarnessSoL-Pi scales auto-research loops for coding agents by optimizing harness-layer mechanisms to reduce token traffic and costs.NVIDIA13 citations137 upvotesRead →An Open Recipe for IMO Gold: Training Nemotron for Olympiad MathematicsA natural-language proof-generation pipeline using post-trained Nemotron 3 Ultra checkpoints achieves gold-medal performance on IMO 2026 through iterative verification and refinement without external tools.NVIDIA43 upvotesRead →Online Draft Co-Training for Speculative Decoding in Large-Scale, Long-Context RL Post-TrainingA system for large-scale online draft co-training accelerates speculative decoding in RL post-training by extending context-parallel attention and adding cross-stage feature transport.NVIDIA4 citations35 upvotesRead →Sol-Attn: Accelerating Video Generation Inference via On-the-Fly Attention SparsificationSol-Attn improves training-free sparse attention for diffusion transformers by combining dynamic block routing, sparse computation, and approximation correction in a single online pass to accelerate video generation without sacrificing quality.NVIDIA11 citations37 upvotesRead →SANA-Video 2.0: Hybrid Linear Attention with Attention Residuals for Efficient Video GenerationSANA-Video 2.0 is a hybrid video diffusion transformer that combines linear and softmax attention to generate high-resolution video efficiently on a single GPU.NVIDIA6 citations38 upvotesRead →Molt: A Scalable PyTorch-Native Training Framework for Agentic Reinforcement LearningMolt is a compact PyTorch framework for agentic reinforcement learning that enables efficient asynchronous training of multimodal and mixture-of-experts policies with minimal overhead.NVIDIA1 citations33 upvotesRead →NVIDIA-labs OO Agents: Native Python Object-Oriented AgentsNOOA treats AI agents as Python objects whose methods and fields define actions, state, and prompts, enabling deterministic testing and LLM-driven runtime completion within a unified programming model.NVIDIA5 citations36 upvotesRead →ASPIRE: Agentic /Skills Discovery for RoboticsASPIRE is a continual learning system that autonomously develops and refines robot control programs through iterative exploration, achieving superior performance and zero-shot generalization in manipulation and household tasks while enabling sim-to-real transfer.NVIDIA114 citations25 upvotesRead →PhysisForcing: Physics Reinforced World Simulator for Robotic ManipulationPhysisForcing enhances embodied video generation by enforcing physical consistency through pixel-level trajectory alignment and semantic-level relational alignment losses in a DiT-based framework.NVIDIA9 citations53 upvotesRead →Causal-rCM: A Unified Teacher-Forcing and Self-Forcing Open Recipe for Autoregressive Diffusion Distillation in Streaming Video Generation and Interactive World ModelsAutoregressive video diffusion extends diffusion distillation frameworks to real-time streaming generation through causal training paradigms, achieving state-of-the-art performance with fast convergence and interactive world modeling capabilities.NVIDIA37 citations28 upvotesRead →Zone of Proximal Policy Optimization: Teacher in Prompts, Not GradientsZone of Proximal Policy Optimization (ZPPO) improves knowledge distillation by using reformulated prompts that help students learn from both correct and incorrect responses, enhancing performance especially at smaller model sizes.NVIDIA4 citations64 upvotesRead →SpatialClaw: Rethinking Action Interface for Agentic Spatial ReasoningSpatialClaw is a training-free framework that uses code as an action interface to enable flexible, stateful spatial reasoning in vision-language models, achieving superior performance across diverse 3D/4D spatial reasoning tasks.NVIDIA23 citations113 upvotesRead →Cosmos 3: Omnimodal World Models for Physical AIOne model that reads and generates text, images, video, audio and robot actions for physical AI.NVIDIA225 citations144 upvotesRead →SANA-Streaming: Real-time Streaming Video Editing with Hybrid Diffusion TransformerSANA-Streaming enables real-time high-resolution video-to-video editing through a hybrid diffusion transformer architecture, cycle-reverse regularization, and efficient system co-design optimized for consumer GPUs.NVIDIA15 citations38 upvotesRead →Why Far Looks Up: Probing Spatial Representation in Vision-Language ModelsVision-language models exhibit entangled spatial representations that correlate vertical image position with distance, impacting reasoning robustness and performance across benchmarks.NVIDIA3 citations53 upvotesRead →Agent Explorative Policy Optimization for Multimodal Agentic ReasoningAgents using vision-language models with extended reasoning face challenges in tool utilization, which are addressed through AXPO, a method that improves performance by optimizing thinking prefixes and tool call resampling.NVIDIA3 citations89 upvotesRead →Gamma-World: Generative Multi-Agent World Modeling Beyond Two PlayersA generative multi-agent world model is presented that uses simplex rotary agent encoding and sparse hub attention to enable scalable, permutation-symmetric interaction between multiple agents in interactive video generation.NVIDIA19 citations148 upvotesRead →LocateAnything: Fast and High-Quality Vision-Language Grounding with Parallel Box DecodingParallel Box Decoding enables efficient and accurate unified visual grounding and detection by decoding geometric elements as atomic units, improving both throughput and localization quality.NVIDIA38 citations142 upvotesRead →PiD: Fast and High-Resolution Latent Decoding with Pixel DiffusionPiD introduces a pixel diffusion decoder that reformulates latent decoding as conditional pixel diffusion, enabling fast and high-quality image synthesis at high resolutions with reduced computational requirements.NVIDIA10 citations45 upvotesRead →Gated DeltaNet-2: Decoupling Erase and Write in Linear AttentionGated DeltaNet-2 improves upon existing linear attention models by separating erase and write operations through distinct channel-wise gates, achieving superior performance in long-context language modeling and retrieval tasks.NVIDIA48 citations30 upvotesRead →LongLive-2.0: An NVFP4 Parallel Infrastructure for Long Video GenerationLongLive-2.0 presents an NVFP4-based parallel infrastructure for long video generation that addresses training and inference bottlenecks through sequence-parallel autoregressive training and diffusion model tuning.NVIDIA37 citations117 upvotesRead →MemLens: Benchmarking Multimodal Long-Term Memory in Large Vision-Language ModelsA new benchmark evaluates memory capabilities in vision-language models through multi-session conversations, revealing limitations of both long-context and memory-augmented approaches.NVIDIA9 citations77 upvotesRead →SANA-WM: Efficient Minute-Scale World Modeling with Hybrid Linear Diffusion TransformerSANA-WM is an efficient 2.6B-parameter world model that generates high-fidelity 720p videos with precise camera control, achieving industrial-level quality while significantly reducing computational requirements through hybrid attention, dual-camera branches, two-stage generation, and robust annotation pipelines.NVIDIA70 citations92 upvotesRead →AnyFlow: Any-Step Video Diffusion Model with On-Policy Flow Map DistillationAnyFlow introduces a novel any-step video diffusion distillation framework that improves upon consistency distillation by optimizing full ODE sampling trajectories through flow-map transition learning and backward simulation techniques.NVIDIA34 citations103 upvotesRead →Maximal Brain Damage Without Data or Optimization: Disrupting Neural Networks via Sign-Bit FlipsDeep neural networks exhibit catastrophic vulnerability to minimal parameter bit flips across multiple domains, which can be identified and mitigated through targeted protection strategies.NVIDIA4 citations57 upvotesRead →Lyra 2.0: Explorable Generative 3D WorldsLyra 2.0 enables large-scale 3D scene creation through persistent video generation that addresses spatial forgetting and temporal drifting issues in long-horizon video models.NVIDIA50 citations42 upvotesRead →Nemotron 3 Super: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic ReasoningNemotron 3 Super is a 120 billion parameter hybrid Mamba-Attention Mixture-of-Experts model pre-trained in NVFP4 with LatentMoE architecture and MTP layers for accelerated inference, achieving superior throughput compared to existing models.NVIDIA54 citations39 upvotesRead →Audio Flamingo Next: Next-Generation Open Audio-Language Models for Speech, Sound, and MusicAudio Flamingo Next represents a significant advancement in audio-language modeling with enhanced understanding capabilities, extended audio input lengths, and novel temporal reasoning mechanisms.NVIDIA40 citations29 upvotesRead →FP4 Explore, BF16 Train: Diffusion Reinforcement Learning via Efficient Rollout ScalingA novel two-stage reinforcement learning framework called Sol-RL integrates FP4 quantization with diffusion model alignment to accelerate training while maintaining high-fidelity performance.NVIDIA3 citations29 upvotesRead →TriAttention: Efficient Long Reasoning with Trigonometric KV CompressionTriAttention addresses KV cache memory bottlenecks in LLMs by leveraging Q/K vector concentration in pre-RoPE space to improve key importance estimation and enable efficient long-context generation.NVIDIA25 citations112 upvotesRead →Nemotron-Cascade 2: Post-Training LLMs with Cascade RL and Multi-Domain On-Policy DistillationNemotron-Cascade 2 is a 30B parameter Mixture-of-Experts model with 3B activated parameters that achieves exceptional reasoning and agentic capabilities, matching frontier open models despite its compact size and demonstrating high intelligence density.NVIDIA63 citations70 upvotesRead →MM-Zero: Self-Evolving Multi-Model Vision Language Models From Zero DataMM-Zero enables zero-data self-evolution of vision-language models through a multi-role framework with proposer, coder, and solver components trained via group relative policy optimization.NVIDIA24 citations55 upvotesRead →Mode Seeking meets Mean Seeking for Fast Long Video GenerationA training paradigm combining mode seeking and mean seeking in a Decoupled Diffusion Transformer enables efficient generation of high-quality long videos by leveraging both global flow matching and local distribution matching techniques.NVIDIA14 citations41 upvotesRead →On Data Engineering for Scaling LLM Terminal CapabilitiesResearchers developed a synthetic task generation pipeline and analyzed data strategies to improve terminal agent performance, creating a large-scale dataset and models that outperform larger counterparts on benchmark tests.NVIDIA44 citations104 upvotesRead →Test-Time Training with KV Binding Is Secretly Linear AttentionTest-time training is reinterpreted as learned linear attention rather than memorization, offering architectural simplifications and improved efficiency.NVIDIA16 citations32 upvotesRead →
About and links
Topics: Multimodal and robotics 31, Inference and efficiency 15, Reasoning 13, Training and scaling 10, Agents and evaluation 9, Foundation models 5, Architectures 4, Alignment and safety 2, Retrieval and data 1
NVIDIA lab profile on Labs (labs.fru.dev): leaders, models, landmarksNVIDIA on Companies (companies.fru.dev)Also onAcquisitions·Quarterly·Paydays·TechConf