News for hackers. Stuff that matters.
Re:Semantic reconstruction of continuous language from non-i... by entwife on Friday September 30, 2022 @07:23PM (#33041994)
The claims in the abstract are inconsistent with what was previously known about the limits of functional MRI. (e.g. [0]). I am skeptical of this result.

[0] https://acnp.org/wp-content/uploads/2017/11/CH26_343-356.pdf Bandettini, PA. "THE SPATIAL, TEMPORAL, AND INTERPRETIVE LIMITS OF FUNCTIONAL MRI" in Neuropsychopharmacology: The Fifth Generation of Progress. Edited by Kenneth L. Davis, Dennis Charney, Joseph T. Coyle, and Charles Nemeroff. American College of Neuropsychopharmacology. 2002.

Re:Semantic reconstruction of continuous language from non-i... by JPLeRouzic on Saturday October 01, 2022 @04:07AM (#33044918)
Since 2002 as a layman I would expect some progress has been achieved?
Re:Semantic reconstruction of continuous language from non-i... by morelisp on Saturday October 01, 2022 @11:55AM (#33047518)
fMRI has fundamental limits due to measuring blood flow as a proxy for neural activity. The resolution remains on the order of millimeters and seconds.
Re:Semantic reconstruction of continuous language from non-i... by sacrosancty on Friday September 30, 2022 @08:56PM (#33042674)
What limitation exactly?

This one uses a language model to predict likely next words in the sequence so it doesn't have to identify an isolated word from the whole dictionary. Considering these models can construct valid English passages from absolutely no information, it's not a stretch to imagine something meaningful being generated from even small hints.