AutoSR: Automatic Symbolic Regression by Searching Research States
We introduce Automatic Symbolic Regression (AutoSR), a fully automated system that instantiates Research-Space Symbolic Regression by searching persistent scientific investigations rather than isolated equations. Finite, noisy data often yield numerically competitive expressions that imply very different behavior outside the observed regime, making numerical fit and syntactic complexity insufficient measures of scientific credibility. Existing approaches largely focus on improving expressions, yet the search typically retains little beyond the resulting formula and score, losing the scientific
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- FuzzySimilar name plus overlapping authors · 67%google-research/google-research →
“Title similarity 0.73; shared authors: sun”
- FuzzySimilar title/name (fuzzy) · 59%wanshuiyin/Auto-claude-code-research-in-sleep →
“Fuzzy title match (0.73): “AutoSR: Automatic Symbolic Regression by Searching Research ” ≈ “wanshuiyin/Auto-claude-code-research-in-sleep””
- LinkedLinked via arxiv author · 85%Kejia Zhang →
“AutoSR: Automatic Symbolic Regression by Searching Research States”
- LinkedLinked via arxiv author · 85%Youran Sun →
“AutoSR: Automatic Symbolic Regression by Searching Research States”
- LinkedLinked via arxiv author · 85%Xinyu Ren →
“AutoSR: Automatic Symbolic Regression by Searching Research States”
- LinkedLinked via arxiv author · 85%Chugang Yi →
“AutoSR: Automatic Symbolic Regression by Searching Research States”
- LinkedLinked via arxiv author · 85%Haizhao Yang →
“AutoSR: Automatic Symbolic Regression by Searching Research States”