Why LLMs can innovate: The role of search and verification
bindureddy · x · 2026-08-31
Core Thesis: LLMs invent new things not because answers are in the training data, but because they master operators and patterns. They innovate by performing massive test-time search, composing known elements, and using verifiers to filter for truth.
Technical Logic:
- Proposal vs. Verification: Proposing solutions is hard, but verifying correctness (e.g., using Lean for math proofs) is cheap.
- Search Space + Verifier: In domains with known rules (biology, protein folding, materials), LLMs can explore vast search spaces and rely on verifiers to find valid results.
Examples:
- AlphaProof: Used Lean as a verifier to search the proof space, winning IMO silver.
- FunSearch: Combined LLM proposals with a checker to discover genuinely new math results for the cap set problem.
- AlphaFold: Generalized to uncrystallized proteins by learning the rules of folding.
Implication: Fields like biology and energy fit the "massive search space + cheap verification" pattern. Once AI learns the universe's patterns, cracking cancer or infinite energy becomes theoretically possible.
Related event: Why LLMs Can Invent New Things: Combinatorial Search and Verification(3 posts)→
More from Research
- Hunyuan Hy4 helps raise lower bound of century-old geometry problem by 8% — teortaxesTex · 2026-08-31
- AI Persona Mechanism: Concept is Fuzzy and Lacks Unified Implementation — voooooogel · 2026-08-31
- The User is not a simple feature or independent entity in models — voooooogel · 2026-08-31
- Ropedia Launches HOMIE Gen2, Betting Robots Need Scaled Human Experience, Not Bigger Models — 机器之心 · 2026-08-31
- Tencent's ElephantBench Probes Epistemic Myopia of LLMs on Long-Tail Knowledge — tencent · 2026-08-31
- Code-as-World: Agentic Discovery of Executable World Representations for Physical Reasoning — MirroS-Lab · 2026-08-31