4 papers indexed here
We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.
Not the right person? Other researchers publish under this name.
Recurrent memory requires learning useful updates, but attention added to assist learning can also compromise length generalization. We study bounded delta attention: causal attention over uncorrected update proposals whose bounded correction enters the recurrent update. Its implementation in TraceRelay combines additi...
Copying short-term memory (STM) into a slower store can preserve state across a context boundary, but persistence alone does not ensure that the retained state influences subsequent memory access. We test this distinction in a Phasor Memory Network (PMNet) using Consolidator, a shared slot-local operator that transform...
This work introduces SchemaMem, an attention-based recurrent memory architecture combining chunk-local attention with a persistent, schema-indexed phase state, and identifies a promising retention--optimization trade-off in schema-indexed recurrence.
Under extreme hyperparameter settings where standard normalizations suffer from gradient divergence, MRSNorm successfully prevents numerical explosion and secures stable optimization trajectories, proposing a fundamental paradigm shift toward phasor-based deep representation learning.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
