Elisa Zhang
0 followers · 147 views
on the atlas — 28
- Path to Astra: critical capabilities and frontier safeguards | OpenAI1 savers
- Google’s Fairwind Program: Cyber defense tools for trusted partners1 savers
- Patterns and problems in multiagent systems \ Anthropic3 savers
- How Claude marks AI-generated content | Claude Help Center1 savers
- Introducing MAI-Cyber-1-Flash inside MDASH | Microsoft AI2 savers
- Security incident disclosure — July 20265 savers
- Epoch Capabilities Index | Epoch AI1 savers
- GPT-Red: Unlocking Self-Improvement for Robustness | OpenAI4 savers
- Inkling: Our Open-Weights Model - Thinking Machines Lab14 savers
- SensorFM: Towards a general intelligence and interface for wearable health data1 savers
- IBM, Red Hat and Palo Alto Networks Expand Project Lightwell to Help Organizations Respond to Software Vulnerabilities - Palo Alto Networks1 savers
- requie/AI-Red-Teaming-Guide: A comprehensive guide to adversarial testing and security evaluation of AI systems, helping organizations identify vulnerabilities before attackers exploit them. ·1 savers
- Cybersecurity Glossary | Abnormal AI1 savers
- Berkeley Vulnerability Initiative · Agentic Vulnerability Coverage Map1 savers
- Frontier AI Cybersecurity Observatory1 savers
- Akrites | Patch the Commons, Together1 savers
- Redeploying Claude Fable 5 \ Anthropic1 savers
- Emergent Cyber Behavior: When AI Agents Become Offensive Threat Actors - Irregular1 savers
- Offense at Scale: How Frontier AI Lowers the Cost of Cyber Attacks - Irregular1 savers
- [2606.28277] Towards Automating Scientific Review with Google's Paper Assistant Tool1 savers
- [2603.05706] Reasoning Models Struggle to Control their Chains of Thought1 savers
- GPT-5.6 Preview System Card1 savers
- The Coming Loop | Armin Ronacher's Thoughts and Writings2 savers
- Elisa's Bookshelf / Curius1 savers
- FIRST Mid-Year Vulnerability Forecast Confirms Historic Surge, Projects ~66,000 CVEs in 20261 savers
- https://cs.stanford.edu/~jsteinhardt/ResearchasaStochasticDecisionProcess.html31 savers
- A global workspace in language models \ Anthropic25 savers
- [2507.19457] GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning4 savers