paper-with-me

Papers

When to Trust the Cheap Check: Weak and Strong Verification for Reasoning

2026-02-19 · Shayan Kiyani, Sima Noorani, George Pappas, Hamed Hassani arxiv

Reasoning with LLMs increasingly unfolds inside a broader verification loop. Internally, systems use cheap checks, such as self-consistency or proxy rewards, which we call weak verification. Externally, users inspect outputs and steer the model through feedback until results are trustworthy, which we call strong verification. These signals differ sharply in cost and reliability: strong verification can establish trust but is resource-intensive, while weak verification is fast and scalable but noisy and imperfect. We formalize this tension through weak--strong verification policies, which decide when to accept or reject based on weak verification and when to defer to strong verification. We introduce metrics capturing incorrect acceptance, incorrect rejection, and strong-verification frequency. Over population, we show that optimal policies admit a two-threshold structure and that calibration and sharpness govern the value of weak verifiers. Building on this, we develop an online algorithm that provably controls acceptance and rejection errors without assumptions on the query stream, the language model, or the weak verifier.

📄 PDF Abstract BibTeX arXiv:2602.17633

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Theoretical Analysis of Weak-to-Strong Generalization

2024-05-25 · Hunter Lang, David Sontag, Aravindan Vijayaraghavan

Strong student models can learn from weaker teachers: when trained on the predictions of a weaker model, a strong pretrained student can learn to correct the weak model's errors and generalize to examples where the teach…

Language ModelingLanguage Modelling

Generalizing Trust: Weak-to-Strong Trustworthiness in Language Models

2024-12-31 · Martin Pawelczyk, Lillian Sun, Zhenting Qi, Aounon Kumar 외

The rapid proliferation of generative AI, especially large language models, has led to their integration into a variety of applications. A key phenomenon known as weak-to-strong generalization - where a strong model trai…

Fairness

Trust Functions: Near-Lossless Weak-to-Strong Generalization by Learning When to Trust the Weak Teacher

2026-05-31 · Arda Uzunoglu, Alvin Zhang, Daniel Khashabi arxiv

Weak-to-strong generalization studies how to improve a strong student using supervision from a weaker teacher when reliable labels are scarce. We view this primarily as a data selection problem, where the key challenge i…

Grounded verification of chemical and materials reasoning: detection is the bottleneck

2026-07-19 · Can Polat, Mustafa Kurban, Erchin Serpedin, Hasan Kurban arxiv

Language models are moving into chemistry and materials discovery workflows, where a wrong molecular formula, space group, or formation energy can silently propagate into downstream decisions. These confabulations hide i…

Weak-to-Strong Generalization via Direct On-Policy Distillation

2026-07-06 · Shiyuan Feng, Huan-ang Gao, Haohan Chi, Hanlin Wu 외 arxiv

Reinforcement learning with verifiable rewards (RLVR) is a powerful recipe for improving language-model reasoning, but it is expensive to repeat on every new strong model because the target model must generate many rollo…

Reinforcement Learning