Researchers introduced AgentMercury, a framework that synthesizes persistent, executable business environments — complete with entities, services, and cross-service rules —…
Researchers developed a quantization framework for deploying large vision-language models on memory-constrained mobile hardware, combining a self-generated training data pipeline…
Researchers developed CLEAR, a technique that uses a lightweight hidden-state gate to continuously control the activation strength of a safety…
Researchers introduced FACET, a framework for generating training tasks for terminal/command-line agents that keeps instructions, solutions, and verifiers aligned by…
A cost-aware evaluation compared large language models against specialized embedding models across 37 text-understanding tasks and found the best LLM…
Researchers developed Inject, Align, and Recover (IAR), a three-stage post-training framework that lets language models answer questions about specific document…
Researchers built PolicyGuide, a system that compiles organizational policies into workflow graphs and uses a proactive verifier at user-turn boundaries…
Researchers built Daedalus-150M, a 150-million-parameter language model architected specifically for efficient CPU inference rather than adapted from a GPU-designed model.…
Researchers introduced Chain-of-Experience (CoE), a test-time learning approach that allows large language models to improve iteratively using self-feedback and environmental…
Researchers introduced FlowEvo, a framework addressing the fact that workflows discovered by LLM agents during task execution are typically thrown…
Researchers ran a controlled experiment placing LLM agents on a simulated social platform to study how peer-ranked feeds shape agent…
Cognitive scientist Gary Marcus disputes Anthropic CEO Dario Amodei’s prediction that AI could cure most human disease within five to…
