OpenAI is moving from a limited pilot to broad deployment of its Trusted Access for Cyber program, introducing GPT-5.4-Cyber — a variant of GPT-5.4 fine-tuned to be cyber-permissive — for thousands of vetted defenders. The post OpenAI Scales Trusted Access…
OpenAI is moving from a limited pilot to broad deployment of its Trusted Access for Cyber program, introducing GPT-5.4-Cyber — a variant of GPT-5.4 fine-tuned to be cyber-permissive — for thousands of vetted defenders. The post OpenAI Scales Trusted Access…
OpenAI is moving from a limited pilot to broad deployment of its Trusted Access for Cyber program, introducing GPT-5.4-Cyber — a variant of GPT-5.4 fine-tuned to be cyber-permissive — for thousands of vetted defenders. The post OpenAI Scales Trusted Access…
OpenAI is moving from a limited pilot to broad deployment of its Trusted Access for Cyber program, introducing GPT-5.4-Cyber — a variant of GPT-5.4 fine-tuned to be cyber-permissive — for thousands of vetted defenders. The post OpenAI Scales Trusted Access…
arXiv:2604.15411v1 Announce Type: new Abstract: The paradigm of agentic science requires AI systems to conduct robust reasoning and engage in long-horizon, autonomous exploration. However, current scientific benchmarks remain confined to domain knowledge comprehension and complex reasoning, failing to evaluate the…
arXiv:2512.05722v2 Announce Type: replace Abstract: Chemical reaction mechanisms are the foundation of how chemists evaluate reactivity and feasibility, yet current Computer-Assisted Synthesis Planning (CASP) systems operate without this mechanistic reasoning. We introduce a computational framework that teaches language models to…
arXiv:2604.15414v1 Announce Type: new Abstract: Continual reinforcement learning must balance retention with adaptation, yet many methods still rely on emph{single-model preservation}, committing to one evolving policy as the main reusable solution across tasks. Even when a previously successful policy is…
arXiv:2604.07055v2 Announce Type: replace Abstract: We give a computer-assisted counterexample to the open question, posed by Rudin, Schapire, and Daubechies in COLT 2012, of whether exhaustive AdaBoost always converges to a finite cycle. The construction is based on a block-product…
arXiv:2604.15416v1 Announce Type: new Abstract: Sign-based optimization algorithms, such as SignSGD, have garnered significant attention for their remarkable performance in distributed learning and training large foundation models. Despite their empirical superiority, SignSGD is known to diverge on non-smooth objectives, which…
arXiv:2604.15461v1 Announce Type: new Abstract: LLM-based simulators offer a promising path for generating complex synthetic data where traditional differentially private (DP) methods struggle with high-dimensional user profiles. But can LLMs faithfully reproduce statistical distributions from DP-protected inputs? We evaluate this…
