AI research papers.

Post-route power recovery changes cell choices under tightly coupled timing, electrical and physical constraints, and the value of each change depends on what it costs to repair its side effects.

Training AI agents with reinforcement learning can be challenging because their tools, context, and decision-making are managed by complex frameworks.

arXiv:2610.04949v1 Announce Type: cross Abstract: Reinforcement learning controllers for underwater vehicles are usually trained against one physics representation and deployed against another, so reported performance do…

arXiv:2610.05235v1 Announce Type: cross Abstract: Surface electromyography (sEMG) is a low-power, cost-effective biosignal for hand-pose estimation and gesture classification.

arXiv:2610.07289v1 Announce Type: cross Abstract: Manual repair of program failures is time-consuming and disruptive for software developers, particularly during the pre-submit phase where test failures occur within cont…

We introduce SWE-Game, a benchmark of 247 tasks grounded in 41 executable reference Godot games spanning 13 gameplay categories in 2D and 3D.

Vision-and-Language Navigation (VLN) has largely focused on a single agent following a single instruction, yet many real-world applications require teams of robots to tackle tasks beyond the capabilities of any individua…

Recent work on language-model adaptation has shown that single models can obtain informative training signals by evaluating their behavior in demonstrationor feedback-augmented contexts, with the help of a teacher networ…

We present WebFovea, a vision-based web agent that placed 2nd in the WebRetriever Challenge 2026 with a final score of 57.0 out of 100.

Agentic systems use large language models (LLMs) to carry out concrete tasks.

Video interactive storytelling enables viewers to actively steer how a video unfolds.

As text-to-SQL systems move beyond demonstrations toward real-world deployment, ambiguity in user questions becomes a primary source of errors.

Pixel-space diffusion models avoid the lossy VAE of latent models, which suggests an advantage on downstream tasks where fine-grained detail matters.

Recently, approaches that leverage human video datasets for robot policy training have become increasingly prevalent.

Methods that recover individual latent variables of the world, from nonlinear ICA to dictionary learning and causal representation learning, anchor the latents to observations through reconstruction, auxiliary supervisio…

Training and evaluating interactive language agents typically requires rich user interactions, yet collecting human feedback is expensive and difficult to scale.

Vision-Language-Action (VLA) models have shown strong generalization in robotic manipulation by leveraging rich representations from pretrained vision-language models.

Skeleton-based reasoning prompting is a promising training-free approach for structuring LLM reasoning, but prior work largely assumes an English-centric setting.

The evaluation of coding agents generally strives to be as realistic as possible.

On-policy distillation (OPD) strengthens language-model reasoning, yet whether students acquire new factual knowledge or compositional skill for multi-step reasoning remains unknown.

Dual-process agents pair a fast policy with a slow deliberative model.

Test-time scaling (TTS) improves the reasoning capabilities of large language models by allocating additional inference computation.

Visual foundation models (VFMs) are increasingly integrated into end-to-end autonomous driving for their powerful representations, yet it remains unclear when these representations improve driving performance.

Dense visual text requires image generators to reproduce long strings across multiple regions with correct placement and legibility.

Memory-augmented reinforcement learning strengthens LLM agents' ability to solve complex long-horizon tasks.

Single-instrument automatic music transcription (AMT) has advanced substantially, yet choral applications require soprano, alto, tenor, and bass (SATB) to be transcribed as separate parts.

Evaluating agents by outcomes alone can obscure the capabilities that produce them.

Prevailing multi-vector visual document retrievers store each page as about a thousand patch vectors, often in vector databases run by a third party.

Product-centric advertisement video generation aims to create promotional videos that preserve fine-grained product identity while presenting selling points through coherent multi-shot narratives.

The architectural design of Large Language Models (LLMs) is shifting from traditional full-attention-only models to hybrid models, which combine different attention modules to improve long-context efficiency and performa…

Large language model agents can improve across tasks by retaining reusable skills distilled from prior interactions.

Large-scale egocentric human interaction data is becoming an important source of physical supervision for embodied learning, yet video alone leaves the contact and pressure that characterize physical interaction unrecord…

Recent advances in multimodal foundation models have made them capable generalist physical agents for a range of manipulation tasks.

General-purpose agents increasingly write code, use tools, and complete complex digital tasks, raising the question of how far these capabilities carry into the physical world.

Autoregressive video generation requires denoising the current frames while writing their key-value representations as context for future predictions.

Flow policies capture rich and diverse action distributions, and fine-tuning them with off-policy RL to improve beyond the demonstrations has drawn growing interest.

Much knowledge work produces new deliverables from files a workspace already holds, and LLM agents are beginning to take such work over.

Multi-teacher on-policy distillation (MOPD) is used in two settings.

We introduce QuadTok, a novel framework for visual tokenization and autoregressive image generation.

Recovering faithful 3D hand motion from video remains challenging due to frequent occlusions and incomplete visual observations, which make frame-wise pose estimates unreliable and temporally inconsistent.

Latent world models have shown a remarkable ability to predict future states and to plan in the real world.

Highly compressed video autoencoders offer an effective way to accelerate video diffusion models, as the Diffusion Transformer (DiT) operates on far fewer tokens.

Real-time robot control demands enough visual history to infer motion and task progress, but processing that history can delay action.

Conditional memory architectures such as DeepSeek Engram use input n-grams to look up learned embeddings, expanding the capacity of large language models (LLMs) with limited additional computation.

We propose Tetris3D, a generative framework for single-image 3D scene reconstruction that recovers objects which are physically and geometrically coherent as a scene.

Language models trained with long-horizon agentic reinforcement learning can generalize knowledge through reasoning, express precise actions, and pursue goals over many steps, raising the ceiling on what an embodied agen…

When audio, video, and text disagree, accuracy alone does not show whether to acquire another source or abstain.

A large language model can only use the text that fits in its context window, and it recomputes its internal key-value (KV) state for a prompt every time the prompt is sent.

Video-language models commonly assume that more temporal observations lead to more reliable reasoning.

Native NVFP4 promises to accelerate reinforcement learning (RL) for large language models (LLMs), but achieving stable learning across separate training and sampling engines remains challenging.