ColliderVM protocol for computation and L2 bridges

[Victor Kolobov]

I would like to update you about an alternative protocol to BitVM that we have been working on, which we call ColliderVM. It uses the same trust assumptions as BitVM but without needing fraud proofs. By that it is eliminating the need for a fraud-proof time window. All in all, some additional work is needed to make this protocol practical.

Perhaps some ideas here could also be of independent interest. 

Here is the link to the paper: 

https://eprint.iacr.org/2025/591 

(Please note that we erroneously used an incorrect estimate for BLAKE3 script size, but this will be fixed in the next version which is coming in a few days):

We’ve also made a video explaining this protocol

https://starkware.co/blog/avihu-levy-bitcoin-horizons-from-op_cat-to-covenants-hong-kong/

Delving Bitcoin post with more details:

https://delvingbitcoin.org/t/collidervm-protocol-for-computation-and-l2-bridges/1662