Inherent Trade-offs when Enforcing Correct Data Symmetries in Multi-head Self-attention
[Submitted on 31 Aug 2026 (v1), last revised 15 Oct 2026 (this version, v2)] View PDF HTML (experimental) Abstract:We give…
[Submitted on 31 Aug 2026 (v1), last revised 15 Oct 2026 (this version, v2)] View PDF HTML (experimental) Abstract:We give…