LOADING INSTRUMENTS…
Generated 09 Sept 2026, 12:41 UTC from indexed data. 101 records in this window. Every item links to its source; counts compare with the preceding window of equal length.
Introducing GPT-6 Astra, our most intelligent and aligned model yet, with state-of-the-art capabilities across computer use, coding, cybersecurity, and science.
OpenAI introduces Daybreak for Frontline Defenders. A $1 billion commitment expands access to frontier cyber AI, training, and support for essential services.
While research on recursive self-improvement (RSI) has predominantly automated model training pipelines, reliable autonomous development demands a missing pillar: post-hoc monitoring and auditing to understand what mode…
As Large Language Model (LLM) agents increasingly automate offensive operations across the cyber kill chain, their efficacy in complex local post-exploitation tasks remains inadequately quantified. Among these, Linux pr…
Training capable cyber agents is often treated primarily as a problem of model scale, yet open-weight post-training is constrained more directly by the cost of executable environments, reliable multi-turn supervision, a…
Large Language Models demonstrate remarkable proficiency in static reasoning, yet training them as autonomous agents through Reinforcement Learning (RL) for long-horizon tasks is often hindered by severe reward sparsity…
Large language model (LLM) agents are increasingly employed for offensive cybersecurity tasks such as automated vulnerability discovery, reconnaissance, and penetration testing. This new capability also threatens one of…
We study scheming in LLM agents, in which agents covertly pursue misaligned goals. Our focus is to understand how scheming arises from the interaction of key factors, such as instrumental goals, environmental affordance…
Social interaction is central to children's language learning, but the effects of different forms of caregiver feedback are difficult to isolate in naturalistic data. We use child-like language models as controlled lear…
Existing methods for test-time reinforcement learning (TTRL) derive rewards from answer-level self-voting on unlabeled test-time tasks with canonical answers, but this breaks down for code generation because programs ca…
Agent harnesses (the system prompt, tool set, execution hooks, and context-management scaffolding around a model) are a critical determinant of agentic task success. Automated harness evolution can enable smaller models…
GPT-6 Astra is our most capable broadly deployed model and our first to reach the Critical level of cybersecurity capability under our Preparedness Framework.
GPT-6 Astra is our most capable broadly deployed model and our first to reach the Critical level of cybersecurity capability under our Preparedness Framework.
DoguLabs tracks publicly available evidence. This brief is assembled from indexed records; no text is generated beyond counts and record fields.
Multilingual LLMs show stereotype-related behavior that varies across languages, but behavioral scores do not show where the relevant information is represented or how it affects the output. To investigate these interna…
Integrating robust safety guardrails into Large Language Models (LLMs) is essential for delivering helpful yet harmless responses. While proprietary systems exhibit reliable safety controls, their underlying methodologi…
Real-world Multi-Objective Reinforcement Learning (MORL) often suffers from sparse rewards, reward conflicts, and late-stage reward tug-of-war, causing traditional linear scalarization to experience severe metric oscill…
Existing methods for test-time reinforcement learning (TTRL) derive rewards from answer-level self-voting on unlabeled test-time tasks with canonical answers, but this breaks down for code generation because programs ca…
We introduce HoneyRoute, an inference-serving layer that detects whether an incoming request is malicious and, if so, routes it to a dedicated honeypot model, shielding production while the adversary's interaction is co…
TL;DR If you train a probe for some property (like "honesty") and do gradient descent against this probe while continuing training that incentivizes dishonesty, the model will change its internal representation to evade…
Long horizon Large Language Model (LLM) agents rely on external memory systems to preserve user preferences and task knowledge across extended interactions. Conventional retrieval mechanisms optimize semantic compatibil…
While research on recursive self-improvement (RSI) has predominantly automated model training pipelines, reliable autonomous development demands a missing pillar: post-hoc monitoring and auditing to understand what mode…
As Large Language Model (LLM) agents increasingly automate offensive operations across the cyber kill chain, their efficacy in complex local post-exploitation tasks remains inadequately quantified. Among these, Linux pr…
As LLMs are increasingly used for pre-submission self-review, there is growing demand for feedback that not only identifies weaknesses but also guides authors toward concrete revisions. We study this as Actionable Peer-…
Reasoning representations are increasingly used as explanations for large language model outputs. Yet they are typically evaluated with model-centric criteria, such as answer accuracy and faithfulness, leaving it unclea…
Large Language Models (LLMs) demonstrate impressive performance across diverse NLP tasks, yet their ability to exhibit genuine contextual understanding remains uncertain. Traditional evaluation metrics such as perplexit…