Verified news, launches and analysis to understand what is changing in artificial intelligence and why it matters to you.
166 articles
A preprint describes a system that changes mechanisms and equations in battery models when their predictions differ from experiments. Its abstract reports results on two tasks, but does not provide enough evidence to assess reproducibility or validity beyond those cases.
Researchers present a method to project the outputs of biomolecular diffusion models, correcting physical defects such as chain clashes or distorted bonds without the need to retrain the model.
A new preprint introduces Long-Transduction, a test designed to isolate how errors in AI models grow as context lengthens, input data formats change, or sub-task complexity increases. The authors report significant performance drops in seven open-weight models, highlighting a critical vulnerability for long-duration agentic workflows.
A recent preprint proposes a system that dynamically adjusts the distribution of workers between prefill and decode phases to prevent SLO violations without requiring additional GPUs.
A preprint introduces decision checkpoints to distinguish whether an agent fails to find a paper, inspect it, or answer correctly after consulting it. The results reveal distinctions hidden by final-answer accuracy, but apply to a specific experimental setting.
A preprint introduces a technique that retrieves continuations from the current trajectory and combines them with drafts generated by another model. The abstract reports improvements in acceptance and throughput but does not provide detailed figures to assess its practical scope.
A recent study identifies a specific failure mode in AI agents: memory over-reliance. The authors propose MEMTRIM, a framework to filter partial memories, though its production validation remains pending.
A technical report introduces a time-series model designed for demand-specific challenges such as frequent zeros and stockouts. It combines a specialised corpus with an adapter attached to a general-purpose model.
A preprint tests whether disagreement among a small model’s answers can predict when a query should be routed to a larger model. Results on GSM8K show an improvement, but the study also flags methodological pitfalls and operational costs.
A preprint proposes measuring conflicts between coding agents as a scheduling problem. Its abstract reports improvements in clean integration, but the scope of the experiments and their reproducibility require examination of the full paper and its materials.
H Company introduces two agent models capable of switching between graphical interfaces, code execution, and MCP or API tools. Weights and results are published, though tests do not prove general autonomy or uniform performance across all environments.
Google DeepMind and GovTech Singapore have announced a tool that uses on-device Gemma models to analyze potential conversational scams. The available information does not specify the model variant, test results or the full scope of its privacy protections.