Test-time scaling has emerged as a key inference strategy for reasoning models, where a mo…
Tool-using language model agents rely on environmental facts to decide whether an action i…
Control evaluations measure whether safeguards can prevent an AI model from causing undesi…
Research on sycophancy asks whether large language models (LLMs) bend towards what their u…
Masked diffusion language models generate tokens in parallel, but their efficiency depends…
Industrial organizations hold operational knowledge in structured records, such as telemet…
Hidden future-token leakage can create false architectural gains: an autoregressive model …
Individual preferences often become visible only after deployment, when a user accepts one…
Large language models (LLMs) are increasingly used to generate domain-specific commercial …
Coordinating two players in a cooking task requires a vision-language agent to track chang…
Recommender systems typically serve one user at a time. However, in households, members sh…
Reusable skills allow large language model agents to package task-specific procedures, too…
When a user says "Keep it short", how short should the LLM's response be? More generally, …
Automated agentic-system design is typically evaluated on static task suites with immediat…
World action models increasingly rely on pretrained video models assumed to capture physic…
This paper studies the problem of safe tool execution for large language model (LLM) agent…
Autonomous research agents are increasingly used to automate large portions of scientific …
The rise of Agentic AI systems has changed the landscape of artificial intelligence by ena…
Tree-based speculative decoding accelerates large language model inference by verifying mu…
GUI agents aim to automate complex user tasks by interacting with mobile interfaces, but e…
Data agents built on ReAct-style loops now execute complex analytical workflows over struc…
Decision models map stated rules and observed states to discrete answers. Aggregate accura…
Harness optimization, which improves an agent's prompts, tools, and workflows, is an incre…
Text skills can extend the capabilities of tool-using agents, but long skill descriptions …
This study introduces GuideEvolve, a self-evolving agent framework incorporating expert gu…
Benchmark performance is the dominant standard for comparing large language models (LLMs),…
Given the enormous number of claims to be checked, verifiable legal AI requires scalable, …
Forming an accurate theory-of-mind (ToM) is key to successful multi-agent collaboration an…
As coding agents become more capable, developers increasingly use them to fix bugs through…
Incorporating AI systems into safety systems requires monitoring to avoid silent faults an…
Literature-searching language agents are becoming a standard component of scientific assis…
Using an LLM to generate training data for another is among the most general ways to train…
Choosing among robot demonstration corpora is a ranking problem: an upstream diagnostic mu…
Parallel thinking has emerged as a promising paradigm for enhancing the reasoning capabili…
We determine the minimum number of directional commands needed to solve alevel of Order of…
A model's failures can reveal gaps in the behavioral criteria guiding its training. We int…
Long-horizon dialogue agents need to keep user memories stable enough for personalization …
World models are increasingly used for planning, where a planner ranks candidate actions b…
Multimodal multi-label emotion recognition is important for affect-aware systems, but mode…
Positional superko (PSK) forbids recreating any earlier whole-board position. It is part o…
Multi-token prediction (MTP) can accelerate language-model decoding by drafting several to…
Modern distributed microservice deployments generate massive volumes of heterogeneous, mul…
Current evaluation of LLM-generated debate largely focuses on argumentative quality, leavi…
Computer-use agents are limited by the interfaces humans give them. An agent may type a sp…
RTL repair benchmarks should measure whether a model can fix broken RTL given a specificat…
Autoregressive models (ARMs) face a "Sequentiality Bottleneck" in reasoning, while diffusi…
In this report, we introduce Sage-A1.5, a 35B Mixture-of-Experts agentic model that builds…
Generative retrieval (GR) casts search and recommendation as autoregressive generation of …
Many scientific phenomena are best understood as outcomes of underlying optimization probl…
Visual anomaly detection aims to detect abnormal objects, while anomaly samples are often …
How can we effectively oversee AI systems that surpass human intelligence? AI Safety via D…
Post-training quantization (PTQ) compresses the weights of a trained model with limited co…
Decision models support routing, guardrails, and output verification in LLM systems. Howev…
AI agents increasingly act on behalf of people or other agents, purchasing services, excha…
Learning latent dynamics is fundamental to building action-conditioned world models for pl…
When a language model develops capabilities not traceable to its individual components, wh…
Human mobility data are fundamental to Web-based location services and broader urban appli…
Agents are starting to sell to other agents. Seller agents write the listings of tools, se…
Large Language Model inference over long contexts is increasingly bottlenecked by the cost…
Recent advances in multimodal large language models (MLLMs) have substantially improved sp…