Safety
Among proponents of neurosymbolic architectures, there had been some debate over the years about whether the outer level would be symbolic (…
Among proponents of neurosymbolic architectures, there had been some debate over the years about whether the outer level would be symbolic (i.e. a harness that calls neural models) or whether the oute
Among proponents of neurosymbolic architectures, there had been some debate over the years about whether the outer level would be symbolic (i.e. a harness that calls neural models) or whether the outer level would be neural (i.e. a neural model using symbolic tool calls as part of its neural inference cycle). Right now, the answer is that the outer level is symbolic -- this is the stronger neurosymbolic formulation. To be exact, it's more like a symbolic sandwich: a symbolic program drives a neural model, but also makes tool calls and writes + runs symbolic scripts as part of its inference cycle. This is a maximalist neurosymbolic vision.
Related
- blast from the past 3.5 years ago; some things have changed (esp. coding and math, via neurosymbolic techniques) but many haven’t:
- Checkmate: you can’t take the harness (which is typically in large part symbolic) away from the neural model without giving up performance. …
- another win for neurosymbolic AI!
Source: Gary Marcus (X) | 2026-08-06