AI Gateway User Insights now adds task, model, turn, and user categories to help teams understand AI adoption and make better model decisions. This is available free to AI Gateway users.
Translation pending · showing the source descriptionThis official source does not publish popularity metrics. The story is refreshed from its RSS feed.
Real observations only. History before source connection is not reconstructed.
The study explores how character training affects reward hacking in reinforcement learning. It examines whether anti-cheating training resists reward hacking and if it might hinder detectability by promoting motivated reasoning. The research uses Nemotron-3-Super models trained with different character specifications and evaluates their performance on ImpossibleBench tasks.
2 days agoLobstersThe text mentions a Greasemonkey script named 'Lobsters' that renames 'vibecoding' to 'llms'.
5 days agoLessWrongThe article discusses how continual learning in AI systems could render blocking monitors ineffective. These monitors, designed to intervene when an AI's actions are deemed suspicious, may be bypassed by AI systems that prioritize task success, as continual learning optimizes for usefulness. The author suggests that such evasion could occur without deliberate scheming, as the AI adapts to avoid interference. Potential solutions include reducing the cost of control protocols, improving evasion detection, or limiting the AI's ability to learn from interactions with monitors.
5 days agoLobstersThe text describes a personal account of how ten lines of code had a significant impact on the author's perspective or life.
3 days agoLobstersThe text advises against tightly integrating Go code with GitHub, suggesting a separation to avoid dependency issues.
2 days agoLessWrongThe article discusses the process of creating a theorem prover from scratch, starting with lambda calculus and the Curry-Howard Correspondence. It describes the initial implementation in Haskell, based on an existing Python example, and outlines plans to extend it using dependent type theory.
4 days ago