Decoder-only transformers fail to base verification decisions solely on current search state in cumulative traces because of scattered retrieval and history entanglement; Selective State Attention enforces state-only decisions via a fixed mask.
Communications of the ACM , volume =
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Can Transformers Learn to Verify During Backtracking Search?
Decoder-only transformers fail to base verification decisions solely on current search state in cumulative traces because of scattered retrieval and history entanglement; Selective State Attention enforces state-only decisions via a fixed mask.