Submissions from ankitmaloo.com

		Testing Anthropic's RSI Claims (ankitmaloo.com)
		1 point by ankit219 36 days ago \| past
		Mythos/Fable-5 is a greedy Depth First Search system (ankitmaloo.com)
		5 points by ankit219 3 months ago \| past
		How Anthropic trained Fable 5 => by analysing its reasoning traces (ankitmaloo.com)
		7 points by ankit219 3 months ago \| past
		LLMs give us a way to factorize intelligence (ankitmaloo.com)
		1 point by ankit219 3 months ago \| past
		World Models (ankitmaloo.com)
		28 points by ankit219 7 months ago \| past \| 4 comments
		Every big lab is putting resources in building world models (ankitmaloo.com)
		2 points by ankit219 7 months ago \| past
		I asked LLM to reverse engineer a unity game. It became a conspiracy theorist (ankitmaloo.com)
		1 point by ankit219 10 months ago \| past
		Using coding LLM agents to hack Catan's browser game (ankitmaloo.com)
		1 point by ankit219 10 months ago \| past
		Using GLM-4.6 to reverse engineer Catan Universe browser game (WebGL and unity) (ankitmaloo.com)
		2 points by ankit219 10 months ago \| past
		RL environments need to account for changing priors to work (ankitmaloo.com)
		2 points by ankit219 11 months ago \| past
		Notes on RL Environments (ankitmaloo.com)
		1 point by ankit219 11 months ago \| past
		Bitter Lesson is about AI agents (ankitmaloo.com)
		139 points by ankit219 on March 23, 2025 \| past \| 105 comments