ceo-bench tests ai agents as startup founders
a new benchmark evaluates language model agents on long-horizon business management by simulating 500 days of running a startup.
aisummaries filed under research
a new benchmark evaluates language model agents on long-horizon business management by simulating 500 days of running a startup.
aigoogle deepmind outlines a control roadmap to manage internal ai agents by treating them as potential insider threats and using layered monitoring.
aia graph neural network study finds adding sparse station data to radar forecasts gives minimal gains, while numerical weather prediction and satellite inputs matter more.
aia new mathematical framework connects shock-wave theory to the learning dynamics of stochastic gradient descent after removing parameter symmetries.
aia new method trains task generators using a lightweight probe instead of costly solver rollouts, making it practical to create frontier tasks for reinforcement learning.
aia statistical theory for offline policy optimization using only trajectory-level outcome labels instead of per-step rewards.
aia new method controls risk in ai agents that retrieve documents and use tools, even when their behavior changes over time.
aia new attention method uses probabilistic routing through learned gaussian components to achieve linear-time sequence mixing, avoiding the quadratic cost of standard attention.
aia new vision-language model pipeline uses closed-loop retrieval and verification to enforce evidence-based outputs and measure step-level faithfulness.
aiz.ai's glm-5.2, a 753b parameter mixture-of-experts model with 1m token context, leads open weights benchmarks and ranks second in code arena webdev.
aimolmomotion predicts future 3d object motion from video, points, and text instructions, outperforming existing methods and aiding robotics and video generation.
aia diffusion-based method jointly models lab values and observation patterns to create realistic irregular clinical time series from mimic-iii data.
aiglm-5.2 is an open-source model with a solid 1m-token context, improved coding, and flexible effort control, ranking highest among open models on long-horizon benchmarks.
aia study compares classification models in the inferbert framework for pharmacovigilance, finding that simpler models can suffice and domain-specific pretraining helps.
aiknowledge edits in multimodal large language models often revert to old facts when image-text pairs are separated, due to modality-specific knowledge storage.
aia study uses mechanistic interpretability and uncertainty quantification to diagnose and repair neural network failures in single-shot 3d imaging at long distances.
aia new bayesian method for boolean matrix factorization provides interpretable binary patterns with uncertainty estimates, applied to copy number analysis in cancer.
ainew concentration inequality for exchangeable sequences gives sharper error estimates when comparing ai model scores on benchmarks.
aia model predictive controller balances production and worker training to handle skill shortages and disruptions in manufacturing supply chains.
aia rule-based agent iteratively creates and refines query rewriting rules to improve bm25 retrieval on legal cases without training.
ai