Category
9,568 papers
Empirical evaluations show that GRASP consistently establishes a new state-of-the-art frontier across diverse datasets, yielding substantial accuracy gains over direct LLM planners on Natural Plan Calendar Scheduling, ZebraLogic, and SciBench Math.
Simulation made broad exploration of models, reasoning settings, and prompts feasible without customer exposure, enabling production improvements that would have been impractical to pursue through live experimentation alone.
PrivDrift is introduced, a benchmark for auditing whether user-disclosed secrets remain recoverable after conversational topic drift and persuasion-based probing, suggesting that privacy risk in active LLM contexts should be evaluated as a persistent behavioral failure mode rather than only as training-data memorizatio...
Self-Play Pretraining with Zero Data is introduced, an initial proof-of-concept towards realizing a more general approach to pretraining, and whether zero-shot performance on natural data improves predictably with self-play compute is tested.
Reach audiences
Advertise in front of researchers, engineers, and readers.
If a language model can recognize code it wrote, it may favor that code as a judge, and instances of one model monitoring each other could collude, and instances of one model monitoring each other could collude.
This work assembles Gold-50, fifty real product and policy episodes whose real-world outcome is known, adjudicated against the public record, and scores the five-way release verdict against it, finding most of the measured gap between frontier cloud models and open-weight models fine-tune and serve offline is attributa...
PUBG Ally is introduced, an embodied agent for PUBG: BATTLEGROUNDS that can reason, act autonomously, and play alongside players as a voice-enabled teammate that combines agentic tool use with real-time game control.
This work constructs a large-scale dataset from over 100K computer science papers, and trains a goal-conditioned reward model to predict citation-impact labels from research goal and idea pairs, and uses this reward to align an idea generator through supervised fine-tuning followed by reinforcement learning.
C3M is presented, a cross-session multimodal memory organization that maintains a bounded active index over persistent source text-image evidence in a bounded, query-blind memory budget for cross-session long-horizon tasks.
The results chart an engineering path toward recursive self-improvement, in which humans distill the principles of model building, agents operationalize them through evidence-driven experimentation, and each generation of AI becomes a more capable architect of the next.
PartHackBench provides a certified control for testing whether evaluator credit changes while all benchmark-defined task-relevant progress remains fixed, and provides a certified control for testing whether evaluator credit changes while all benchmark-defined task-relevant progress remains fixed.
Large language models (LLMs) are increasingly deployed as multi-turn agents that must sustain goals, use tools, and adapt to other agents over extended interactions. However, existing research lacks auditable, multi-turn, multi-factorial experiments that quantify LLM behavior under explicit constraints, with time-resol...
Natural Language Processing
Our most capable vision-language-action model enhances robot dexterity, reliability, and safety
A Blog post by Ai2 on Hugging Face
A Blog post by IBM Research on Hugging Face
Flint-interactive gives AI agents a language for creating interactive visualizations. Click a chart, filter data, highlight related content, and explore beyond the view the author originally chose. The post Flint-interactive: A language for creating interactive data visualizations appeared first on Microsoft Research.
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
