LLM reasoning / chain-of-thought

HRM

Superseded baseline#47 of 772 most-superseded

Superseded — cited as a baseline and beaten by newer methods

1 papers critique it · 1 beat it on benchmarks

What papers say

Verbatim critique sentences, each from a paper that cites HRM as a baseline.

SR² surpasses HRM by 11.6% on Sudoku-Extreme and 19.2% on Maze-Hard, while using only one eighth of HRM's parameters.
Selection, Reflection and Self-Refinement: Revisit Reasoning Tasks via a Causal Lens

Beaten on benchmarks

Head-to-head results where a newer method reports beating HRM. Values are copied from the source paper's tables — verify against the cited paper.

What to use instead

Recent methods in the same sub-problem, not yet superseded in the knowledge base — arXiv benchmark leaders, not vetted production recommendations.