Format Online paper / PDF "Engineers have spent decades developing systems that flag observed deviations from learned normal behavior." View Source Location PART 3 - THE THIRD AXIS CHAPTER 15: DEPLOY NOW, PRAY LATER Crossing the Bounds Source title An Intrusion-Detection Model Source authors Dorothy E. Denning Source year 1987 Confidence high Importance mid
Format Online paper / PDF "The model can still produce an output, sometimes with spectacular confidence, despite encountering unfamiliar inputs." View Source Location PART 3 - THE THIRD AXIS CHAPTER 15: DEPLOY NOW, PRAY LATER Crossing the Bounds Source title Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Images Source authors Anh Nguyen, Jason Yosinski, Jeff Clune Source year 2015 Confidence high Importance high
Format Online paper / PDF "A familiar category encountered against an unusual background or under different conditions can constitute a covariate shift." View Source Location PART 3 - THE THIRD AXIS CHAPTER 15: DEPLOY NOW, PRAY LATER Crossing the Bounds Source title Dissecting Out-of-Distribution Detection and Open-Set Recognition: A Critical Analysis of Methods and Benchmarks Source authors Hongjun Wang, Sagar Vaze, Kai Han Source year 2024 Confidence high Importance mid
Format Online paper / PDF "A model learns statistical regularities from training data, but its past performance provides weaker evidence when deployment conditions change." View Source Location PART 3 - THE THIRD AXIS CHAPTER 15: DEPLOY NOW, PRAY LATER Crossing the Bounds Source title Understanding Failures in Out-of-Distribution Detection with Deep Generative Models Source authors Lily H. Zhang, Mark Goldstein, Rajesh Ranganath Source year 2021 Confidence high Importance high
Format Online paper / PDF "When deployment conditions stop resembling the past, the problem is distribution shift, with out-of-distribution inputs being one important part of it." View Source Location PART 3 - THE THIRD AXIS CHAPTER 15: DEPLOY NOW, PRAY LATER Crossing the Bounds Source title Dissecting Out-of-Distribution Detection and Open-Set Recognition: A Critical Analysis of Methods and Benchmarks Source authors Hongjun Wang, Sagar Vaze, Kai Han Source year 2024 Confidence high Importance high
Format Online paper / PDF "Automation can leave humans responsible for abnormal conditions while diminishing their opportunities to practice the skills needed to intervene." View Source Location PART 3 - THE THIRD AXIS CHAPTER 15: DEPLOY NOW, PRAY LATER How Code Diverges From Prose Source title Ironies of Automation Source authors Lisanne Bainbridge Source year 1983 Confidence high Importance high
Format Online paper / PDF "Apple researchers led by Parshin Shojaee observed a complete accuracy collapse in tested reasoning models beyond certain puzzle-complexity thresholds." View Source Location PART 3 - THE THIRD AXIS CHAPTER 15: DEPLOY NOW, PRAY LATER How Code Diverges From Prose Source title The Illusion of Thinking: Understanding the Strengths and Limitations of Reasoning Models via the Lens of Problem Complexity Source authors Parshin Shojaee; Iman Mirzadeh; Keivan Alizadeh Vahid; Maxwell Horton; Samy Bengio; Mehrdad Farajtabar Source year 2025 Confidence high Importance mid
Format Online paper / PDF "Models can generate what Zhao and colleagues called “fluent nonsense”, plausible reasoning chains containing logical inconsistencies." View Source Location PART 3 - THE THIRD AXIS CHAPTER 15: DEPLOY NOW, PRAY LATER How Code Diverges From Prose Source title Is Chain-of-Thought Reasoning of LLMs a Mirage? A Data Distribution Lens Source authors Chengshuai Zhao et al. Source year 2026 Confidence high Importance high
Format Online paper / PDF "Fine-tuning can restore performance on newly introduced data distributions without establishing unrestricted reasoning generalization." View Source Location PART 3 - THE THIRD AXIS CHAPTER 15: DEPLOY NOW, PRAY LATER How Code Diverges From Prose Source title Is Chain-of-Thought Reasoning of LLMs a Mirage? A Data Distribution Lens Source authors Chengshuai Zhao et al. Source year 2026 Confidence high Importance mid
Format Online paper / PDF "Zhao and colleagues interpreted their findings as evidence that chain-of-thought can behave like structured pattern matching rather than generalizable logical inference." View Source Location PART 3 - THE THIRD AXIS CHAPTER 15: DEPLOY NOW, PRAY LATER How Code Diverges From Prose Source title Is Chain-of-Thought Reasoning of LLMs a Mirage? A Data Distribution Lens Source authors Chengshuai Zhao et al. Source year 2026 Confidence high Importance high
