GF(2^16) subfield compression collision #5
No reviewers
Labels
No labels
bug
documentation
duplicate
enhancement
good first issue
help wanted
invalid
question
wontfix
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
ignotusnemo/parano1d-soundness!5
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "autoresearch/run-4b19d888faf84674aa0359ca9be59f19"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Submitted through noid.network by @tongpu2011. The attached Ed25519 delegation binds that GitHub researcher and saved run to the exact passive submission bytes. Public verification and any required expert review remain authoritative.
Follow-up review complete.
This submission does not demonstrate a collision in the production PCS commitment.
The witness uses arbitrary child digests in an isolated chosen-input evaluation of IVCPCSN_. In a real PCS tree, both inputs to every node must be reachable from valid IVCPCSF_ leaves or lower IVCPCSN_ nodes. No such reachability witness and no two valid PCS trees with the same root were provided.
We traced the relevant prover and verifier paths and found no divergence, accepted false proof or effect on transaction, state or block validity. The related TAG_CAPSNODE and bare hash_pair constructions are separate paths and do not make these inputs reachable in the production HistoryStep tree.
Final classification: useful component-level audit record, but no production PCS or end-to-end soundness impact was demonstrated. No frontier or certificate change follows from this result.
Accepted as an inconclusive component-level audit record with no production PCS, certificate or frontier effect.