AI research papers.

Bengali, spoken by over 230 million people, remains severely under-resourced in spoken language research.

Learning velocity fields directly in pixel space is difficult and slow.

Are concepts and mechanisms unified across modalities in current AI models, or instead implemented separately for each one?

Developers download quantized large language models (LLMs) from public registries to run them locally.

Reinforcement learning (RL) has emerged as the primary paradigm for optimizing Large Language Model (LLM) agents, yet efficient critic-free algorithms (e.g., GRPO) suffer from a fundamental uniform credit assignment dile…

Continual learning is the ability of LLM agents to improve over time by acquiring, refining, and applying knowledge from experience while retaining useful prior knowledge.

People have long dreamed of an assistant that is always there to guide them and remind what they forget, from new dishes to lost keys.

Memory is increasingly used in long-horizon agents, yet the net value of building memory over direct retrieval from raw history remains largely unclear.

While Multimodal Large Language Models (MLLMs) excel at zero-shot reasoning, their spatial capabilities are primarily evaluated through static, passive observation.

Extending Vision-Language-Action models (VLAs) to humanoid robots requires coordinated whole-body control and diverse demonstrations that are costly to collect in real-world scenarios.

In this paper, we introduce Δτ-Bench, an executable agent benchmark that models the structural evolution of large tool ecosystems.

Making an open-weight language model generate text is not the same as making it finish coding work.

Training a spatial VLM to answer which side an object lies on does not require it to preserve the metric displacement that determines that answer, while supervision for a distance estimate may leave categorical relations…

Language agents increasingly act through external tools and interactive environments, making the interface between the model and its executable capabilities part of the decision process.

Scaling vision-language-action (VLA) models is constrained by scarce action-labeled demonstrations and incompatible control spaces across robot embodiments.

Recursive self-improvement (RSI) requires an agentic system to design, train, and evaluate its own successor.

Spatial audio authoring coordinates source decisions with soundfield synthesis.

Multi-agent systems often rely on predefined collaboration structures, limiting their ability to adapt to diverse task requirements.

Generative models increasingly expose a reasoning effort setting that lets users scale inference-time compute, trading higher cost for potentially improved performance.

Agent harnesses enable large language models to act through tools, observe the results, and iteratively work toward a task goal.

Reusable Web-agent verification requires a rule to follow changing requirements and support subsequent execution.

Animals have a remarkable capacity for autonomy---driven by intrinsic motivation---without external sensory cues or extrinsic objectives, yet the computational mechanisms behind this ability largely remain a mystery.

While progress in informal mathematics has come largely from scaling the chain-of-thought and agentic reasoning of frontier language models, formal theorem proving has increasingly relied on human-designed workflows arou…

Localized AI manipulations in long-form videos often affect brief events within otherwise authentic content, making video-level classification insufficient for forensic analysis.

Learning visual design is essential for prompt enhancement (PE) to turn underspecified user requests into coherent visual specifications for text-to-image (T2I) generation.

Patent examination is a challenging real-world task that requires agents to discover sparse prior art, iteratively refine queries, inspect technical evidence, and produce novelty assessments documented in Office Actions…

Rubric-based rewards enable effective reinforcement learning on non-verifiable tasks.

Successful tool-use trajectories are not necessarily useful supervision.

Hallucination detection is increasingly important for evaluating and deploying large language models.

Can AI agents autonomously train useful language models and improve them using feedback from users?

GUI grounding requires locating small controls in cluttered screenshots and identifying which control a natural-language instruction refers to.

Many robot manipulation videos do not indicate whether the task was successfully completed.

Generative AI and large language models (LLMs) are moving into the Internet of Things, yet most deployments assume cloud connectivity, and many IoT settings have intermittent, restricted, or no connectivity.

For agentic data analysis, language-model agents often require a large number of annotated tasks to learn how to interact effectively with executable environments (i.e., databases).

Can LLM agents carry a hard real-world problem from raw data to a system that holds its own against people?

Repeated attempts at interactive tasks uncover facts, useful operations, and failures that can inform subsequent execution.

Catastrophic forgetting in continual learning (CL) manifests not only as accuracy degradation but also as explanation drift: models maintain predictive accuracy while silently shifting attention from diagnostic features…

In September 2026, a new large language model was released.

arXiv:1803.03407v3 Announce Type: replace Abstract: Artificial Intelligence (AI) started out with the ambition to reproduce the human mind, but as soon as the sheer scale of that ambition became apparent, it retreated in…

arXiv:2606.08417v3 Announce Type: replace-cross Abstract: Diffusion language models (DLMs) have emerged as the leading non-autoregressive alternative to language modeling.

arXiv:2610.15884v3 Announce Type: replace Abstract: False retry permission (FRP) occurs when an agent retries a timed-out tool call without support from the visible contract.

arXiv:2610.21614v2 Announce Type: replace-cross Abstract: Large audio language models (LALMs) have made substantial progress in audio understanding.

arXiv:2610.30822v1 Announce Type: new Abstract: The increasing adoption of generative AI presents new opportunities while raising AI governance and regulatory challenges, particularly for small and medium-sized enterpris…

Standard sequence models encode position by token rank rather than physical coordinate, typically forcing practitioners to destructively resample continuous physical signals such as spectra observed across heterogeneous…

Developing general-purpose algorithms with global asymptotic convergence guarantees for Nash equilibria of general-sum games remains a longstanding challenge.

Robot foundation models are typically evaluated through costly closed-loop rollouts, motivating low-cost offline proxies for model capability.

Deploying language agents in real applications demands more than the right answer: an agent must execute a multi-step business process correctly, follow domain-specific rules, and interact with a user across a dialogue,…

Recent test-time training (TTT) frameworks have enabled open-weight large language models (LLMs) to outperform inference-only frameworks powered by frontier models on scientific discovery tasks.

Behavior cloning for robotic manipulation is highly sensitive to the coverage of its training data, yet collecting diverse successful demonstrations remains costly.

Humans reason about the physical world by flexibly combining different forms of inference as a problem unfolds.