Robostral Navigate
Deploying navigation systems at scale requires a recipe that minimizes sensor assumptions, generalizes across robot embodiments, and trains efficiently. Yet, today's best systems d...
Research Papers
Latest AI/ML research from ArXiv and Papers with Code
ReferTrack: Referring Then Tracking for Embodied Visual Tracking
Embodied visual tracking (EVT) requires a mobile agent to continuously follow a specific target described in natural language using only onboard vision. While recent vision-languag...
Molt: A Scalable PyTorch-Native Training Framework for Agentic Reinforcement Learning
Agentic reinforcement learning research is constant algorithm modification, new estimators, new pipeline stages, new rollout schemes, and in mainstream frameworks each change threa...
Multimodal Speaker Verification as a Threat to Speaker Anonymization
Most automatic speaker verification (ASV) systems operate on individual utterances, despite real-world interactions typically consisting of multiple utterances. As speech accumulat...
How Fast Can Reward Models Score? A Systems Study of C++ and PyTorch Inference Runtimes for RLHF
In RLHF pipelines, reward scoring blocks policy updates. Slow scoring bottlenecks the entire loop, since no update runs until every rollout gets a score. And yet most setups just d...
Beyond Relevance-Centric Retrieval: Rubric-Oriented Document Set Selection and Ranking
As large language models and AI agents become the primary consumers of search results, document set quality determines the upper bound of downstream generation. Yet existing evalua...
Progress Reward Modeling for Robotic Learning: A Comprehensive Survey
Robotic learning takes place in dynamic environments with large behavior spaces. A terminal success signal only tells the robot whether the task is completed. It does not explain w...
SLAI T-Rex: Full-Parameter Post-training of the DeepSeek-V4 Family on Ascend SuperPOD
Full-parameter post-training of trillion-parameter-scale MoE models introduces substantial system-level challenges for large-scale distributed training, including severe memory pre...
SLPO: Scaling Latent Reasoning via a Surrogate Policy
Reinforcement learning with verifiable rewards has become the predominant recipe for eliciting test-time scaling in explicit Chain-of-Thought reasoners. Yet this scaling path remai...
DocOps: A Verifiable Benchmark for Autonomous Agents in Complex Document Operations
As autonomous agents rapidly evolve, their ability to reliably manipulate ubiquitous digital documents has become critical for enabling general-purpose AI assistants and automating...
LLMs Get Lost in Evolving User Intent
As LLMs become more capable, they are increasingly deployed as collaborative agents, taking on user-delegated tasks through iterative interaction. Yet genuine interaction is inhere...
Train the Model, Not the Reader: Decodability Supervision for Verifiable Activation Explanations
Natural-language autoencoders score explanations of hidden activations by reconstruction: an explanation is deemed faithful if the activation can be regenerated from it. The test i...
ENTRAP-VL: A Taxonomic Probe for Dual Contextual Entrainment in Vision-Language Models
Contextual entrainment is the tendency of a model to let auxiliary context in its input pull its output, independently of whether that context is relevant, true, or even meaningful...
NVIDIA-labs OO Agents: Native Python Object-Oriented Agents
Traditional agent development is split across prompt templates, tool schemas, callback code, and workflow graphs. We present NVIDIA Object-Oriented Agents (NOOA), a model-agnostic ...
Self Gradient Forcing: Native Long Video Extrapolation
Recent autoregressive video diffusion methods are increasingly built upon Self Forcing, where the student is trained on histories produced by its own rollout rather than ground-tru...
ISO: An RLVR-Native Optimization Stack
Reinforcement learning with verifiable rewards (RLVR) is rapidly advancing the reasoning capabilities of language models, yet the optimization layer that converts reward feedback i...
ABot-World-0: Infinite Interactive World Rollout on a Single Desktop GPU
We present ABot-World-0, an action-conditioned video world model for real-time, long-horizon closed-loop interaction, supported by a multi-source data infrastructure spanning AAA g...
Generative World Renderer at the Speed of Play
Generative world renderer AlayaRenderer receives structured world states exported from physics engines and synthesizes RGB frames. Unlike models that generate frames from text/cont...
AutoIndex: Learning Representation Programs for Retrieval
We present AutoIndex, a framework for learning representation programs: executable transformations that map raw documents into the representations exposed to a retrieval system. Ra...
HPD-Parsing: Hierarchical Parallel Document Parsing
Efficient teamwork typically combines global coordination with parallel execution, a principle not yet fully reflected in unified Vision-Language Model (VLM)-based document parsers...
Two-Level Meta-Rubrics for Evaluating Open-Ended Generation: GAMUT, a Benchmark for Factual Completeness
Evaluating the factuality of long-form generations has focused predominantly on precision, measuring whether the claims a model makes are correct. The dominant decompose-search-ver...
Stale but Stable: Staleness-Adaptive Trust Regions for Stabilizing Asynchronous Reinforcement Learning
Asynchronous reinforcement learning improves throughput by decoupling rollout generation from optimization, but staleness is an inevitable byproduct compounded by policy lag, engin...
Computational Humor with Multimodal LLMs: Methods, Datasets, Evaluation, and Challenges
Multimodal humor in memes, cartoons, and comics remains difficult for AI systems because intended meaning depends on non-literal mechanisms, shared cultural knowledge, and communic...
Where Should Optimizer State Live? Tiered State Allocation for Memory-Efficient Mixture-of-Experts Training
Optimizer state is the largest single line item in the memory budget of mixture-of-experts (MoE) training: on a 6.78B-parameter MoE language model, AdamW keeps 50.6 GB of first and...