Amanda Askell | LinkedIn

I work on aligning and evaluating AI systems, with a focus on large language models. I… · Experience: Anthropic · Education: New York University · Location: San Francisco Bay Area · 418 connections on LinkedIn. View Amanda Askell’s profile on LinkedIn, a professional community of 1 billion members.

Arnoldo Frigessi University of Oslo • 2K followers New paper! Available benchmarks for long-context reasoning in Large Language Models do not distinguish the effect of task complexity, presence of irrelevant information, and task length. @daniel Kaiser invented CogniLoad, a synthetic benchmark grounded in Cognitive Load Theory. CogniLoad is used to test 22 reasoning LLMs, and reveals distinct performance sensitivities, identifying task length as a dominant factor, varied tolerances to intrinsic complexity and a never seen before U-shaped response to presence of distracting information.…

saved by

• Hardeep Gambhir

related reading

• Amanda Askellen.wikipedia.org
• Aman's AI Journal • Primers • Ilya Sutskever's Top 30aman.ai
• User awareness in frontier modelstransluce.org
• Discovering Language Model Behaviors with Model-Written Evaluations — LessWronglesswrong.com
• Automated Alignment Researchers: Using large language models to scale scalable oversight \ Anthropicanthropic.com
• [2604.16812] Introspection Adapters: Training LLMs to Report Their Learned Behaviorsarxiv.org
• Shreya Shankarshreya-shankar.com
• Searching for Amanda Askell with Chinese Characteristicscaithrin.com
• [2510.13786] The Art of Scaling Reinforcement Learning Compute for LLMsarxiv.org
• Auditing language models for hidden objectives \ Anthropicanthropic.com
• Self-Adapting Language Modelsarxiv.org
• [2605.21739] AttuneBench: A Conversation-Based Benchmark for LLM Emotional Intelligencearxiv.org