Research
A Verifiable Search Is Not a Learnable Chain-of-Thought
可验证搜索不是可学习的链式思维
A Verifiable Search Is Not a Learnable Chain-of-Thought
arXiv.orgIt is tempting to assume any task solvable by a short program can be taught to a model as its chain-of-thought: write the steps out, fine-tune, and the model follows. This paper shows the assumption fails for an identifiable class of procedures. The testbed is nine reasoning tasks, each from a deterministic generator; public and hidden splits share generators, so held-out data proxies test accuracy. I reverse-engineer the generators into Python solvers, render them as chain-of-thought, and distill into a rank-<= 32 LoRA over a 30B (3.5B-active) Nemotron model. Forward-computable tasks install readily: lookup/arithmetic and an 8-bit boolean task transfer (>= 0.99 and 0.68). Cryptarithm does not: distilling its backtracking search holds at 0.01-0.07 across eleven chain-of-thought designs, RL from verifiable rewards, and self-training, even though a search solver answers 71% of instances. This is not a capability gap. The model does the arithmetic on 97-100% of lines and ranks the correct cipher in its top eight on 71%; it cannot carry the search forward as a left-to-right derivation. Fine-tuning learns the shape of a verifiable elimination step while its verdicts become unconditional templates, correct only 16-57% of the time ("verdict-as-token"). The ceiling holds across backbones from 3B to 671B and across fine-tuning and prompting; a controlled intervention isolates the cause: revealing the cipher key, which turns the derivation forward, lifts the same instances from 0.03 to 0.57. When a procedure's only solution is search over information-free structure, no faithful forward chain-of-thought exists to imitate. The task becomes learnable only by removing the search, precomputing its combinatorial core into a catalog and reducing the trace to recall plus verification; the 1st-place solution reaches Private LB 0.92 this way. What distills is memorization and verification, not search.
Open sourceRecommended because
This is worth tracking because it is a concrete research signal, not just a passing headline. The source preview points to a research result, method, evaluation, dataset, or safety finding. For builders and operators, "A Verifiable Search Is Not a Learnable Chain-of-Thought" can be used as a checkpoint for technical due diligence, roadmap bets, agent design, and evaluation strategy. I keep this thread indexed so future searches around AI research papers, technical methods, and applied AI systems can land on a source-linked page instead of disappearing into a fast-moving feed from arXiv.org.
What to take from this signal
Context
"A Verifiable Search Is Not a Learnable Chain-of-Thought" is archived here as a source-linked AI signal from arXiv.org. The useful part is the connection between Verifiable, Search, Learnable, Chain-of-Thought, tempting and technical due diligence, roadmap bets, agent design, and evaluation strategy, which makes the item more actionable than a normal feed headline. The source context says: It is tempting to assume any task solvable by a short program can be taught to a model as its chain-of-thought: write the steps out, fine-tune, and the model follows. This paper shows the assumption fails for an identifiable class of procedures. The testbed is nine reasoning tasks, each from a deterministic generator; public and hidden splits share generators, so held-out data proxies test accuracy. I reverse-engineer the generators into Python solvers, render them as chain-of-thought, and distill into a rank-<= 32 LoRA over a 30B (3.5B-active) Nemotron model. Forward-computable tasks install readily: lookup/arithmetic and an 8-bit boolean task transfer (>= 0.99 and 0.68). Cryptarithm does not: distilling its backtracking search holds at 0.01-0.07 across eleven chain-of-thought designs, RL from verifiable rewards, and self-training, even though a search solver answers 71% of instances. This is not a capability gap. The model does the arithmetic on 97-100% of lines and ranks the correct cipher in its top eight on 71%; it cannot carry the search forward as a left-to-right derivation. Fine-tuning learns the shape of a verifiable elimination step while its verdicts become unconditional templates, correct only 16-57% of the time ("verdict-as-token"). The ceiling holds across backbones from 3B to 671B and across fine-tuning and prompting; a controlled intervention isolates the cause: revealing the cipher key, which turns the derivation forward, lifts the same instances from 0.03 to 0.57. When a procedure's only solution is search over information-free structure, no faithful forward chain-of-thought exists to imitate. The task becomes learnable only by removing the search, precomputing its combinatorial core into a catalog and reducing the trace to recall plus verification; the 1st-place solution reaches Private LB 0.92 this way. What distills is memorization and verification, not search.
Builder takeaway
For an AI builder, the main takeaway is to watch how this signal changes practical decisions around technical feasibility, evaluation design, safety limits, and product primitives. It can inform what to test next, which product surface to compare, and whether the underlying workflow is ready for real users.
Source context
arXiv.org remains the authoritative source for the original claim. This page adds a stable archive URL, a short builder interpretation, and related search language so the item can be found later when the original feed has moved on.
Search angles
- A Verifiable Search Is Not a Learnable Chain-of-Thought Research context
- arXiv.org AI research
- Verifiable, Search, Learnable, Chain-of-Thought, tempting builder takeaway
- AI research papers, technical methods, and applied AI systems
This page keeps a source preview and a stable archive URL for search discovery. The original source remains authoritative.