Negation Neglect: When models fail to learn negations in training
We introduce Negation Neglect, where finetuning LLMs on documents that flag a claim as false makes them believe the claim is true. For example, models are finetuned on documents that convey "Ed...
Cory Doctorow: Stories Are a Fuggly Hack - Locus
I've mentioned before, stories are weird. I mean, really, really weird. Nothing that happens in a piece of fiction has any consequence in the real world. Romeo and Juliet did not live, did not die, and the ''tragedy'' they represent is objectively less important than the tragedy of the live yogurt culture I digested at the breakfast table this morning. My theory is that the parts of our brains that …Read More
How Learning Happens | Seminal Works in Educational Psychology and Wha
How Learning Happens introduces 28 giants of educational research and their findings on how we learn and what we need to learn effectively, efficiently, and
The NeverEnding Story (1984)
While hiding from bullies in his school's attic, a young boy discovers the extraordinary land of Fantasia, through a magical book called The Neverending Story. The book tells the tale of Atreyu, a…
A field guide to Claude Fable 5: Finding your unknowns | Claude | Claude by Anthropic
Practical patterns for agentic coding with Claude Fable: how to find your unknowns before, during, and after implementation, from the team at Anthropic.
On AI as model organism for human learning
I've a pretty taste for paradox
Neural Networks: Zero To Hero
Master 4 Layers of Storytelling to Make Your Art Unforgettable
You think storytelling is just for writers? Wrong. The story you craft—not just on canvas, but in your journey, your impact, and in your own mind—could make your art impossible to ignore.
Fleetwood on Twitter / X
Studying continual learning at the moment, best papers thus far:https://t.co/Y9oXBAiyj2https://t.co/ByWlaF3ncnhttps://t.co/hG0XIzq6cHhttps://t.co/5VSEnBIkX2— Fleetwood (@fleetwood___) April 11, 2026
MATT THOMPSON
As part of my EngD project, I am creating ways for story authors to easily write narratives with multiple branches, along the lines of the Choose Your Own Adventure stories from the 1990s. Typically, this kind of story authoring is hard, as the number of material that needs to be written increases exponentially with the number of decision points.
The Learning Dispatch | Carl Hendrick | Substack
Making sense of how learning happens. https://www.carlhendrick.com/. Click to read The Learning Dispatch, by Carl Hendrick, a Substack publication with tens of thousands of subscribers.
Ever thought we acquire generalizable knowledge by discarding details and compressing our experiences? In a new BBS paper, @sabinasloman.bsky.social and I argue otherwise, proposing a novel way of studying human learning inspired by double descent in ML. Disagree? Propose a commentary by May 15 :)
