arXiv:2610.41089v1 Announce Type: new
Abstract: Estimating whether a problem is too complex for a large language model (LLM) before the model attempts it can guide model routing, how to allocate test-time compute, and suggest the need for new tools to aid the agent. While existing estimators rely on the problem text, LLM-annotated task difficulty labels, or signals obtained after a model attempts to solve the problem, such as its hidden states, none explicitly leverage the constraint structure inherent in the problem: the variables to assign and how the constraints over them interact.
We introduce CIRCE, which extracts a problem's structure by using a cheap LLM to construct a Constraint Intermediate Representation that deterministically compiles into the constraint graph. Across seven reasoning benchmarks (including SAT, logic, and math) and four LLMs, even a randomly initialized GNN encoder improves predictions of LLM response correctness, reducing negative log-likelihood from 0.465 to 0.423 and increasing AUROC from 0.691 to 0.727 relative to the strongest baseline for each metric.
Abstract: Estimating whether a problem is too complex for a large language model (LLM) before the model attempts it can guide model routing, how to allocate test-time compute, and suggest the need for new tools to aid the agent. While existing estimators rely on the problem text, LLM-annotated task difficulty labels, or signals obtained after a model attempts to solve the problem, such as its hidden states, none explicitly leverage the constraint structure inherent in the problem: the variables to assign and how the constraints over them interact.
We introduce CIRCE, which extracts a problem's structure by using a cheap LLM to construct a Constraint Intermediate Representation that deterministically compiles into the constraint graph. Across seven reasoning benchmarks (including SAT, logic, and math) and four LLMs, even a randomly initialized GNN encoder improves predictions of LLM response correctness, reducing negative log-likelihood from 0.465 to 0.423 and increasing AUROC from 0.691 to 0.727 relative to the strongest baseline for each metric.
This article has been indexed from cs.AI updates on arXiv.org