paper-with-me

Papers

Hard2Verify: A Step-Level Verification Benchmark for Open-Ended Frontier Math

2025-10-15 · Shrey Pandit, Austin Xu, Xuan-Phi Nguyen, Yifei Ming, Caiming Xiong, Shafiq Joty arxiv

Large language model (LLM)-based reasoning systems have recently achieved gold medal-level performance in the IMO 2025 competition, writing mathematical proofs where, to receive full credit, each step must be not only correct but also sufficiently supported. To train LLM-based reasoners in such challenging, open-ended settings, strong verifiers capable of catching step-level mistakes are necessary prerequisites. We introduce Hard2Verify, a human-annotated, step-level verification benchmark produced with over 500 hours of human labor. Hard2Verify is designed to rigorously assess step-level verifiers at the frontier: Verifiers must provide step-level annotations or identify the first error in responses generated by frontier LLMs for very recent, challenging, and open-ended math questions. We evaluate 29 generative critics and process reward models, demonstrating that, beyond a few standouts, open-source verifiers lag closed source models. We subsequently analyze what drives poor performance in step-level verification, the impacts of scaling verifier compute, as well as fundamental questions such as self-verification and verification-generation dynamics.

📄 PDF Abstract BibTeX arXiv:2510.13744

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

The Hidden Signal of Verifier Strictness: Controlling and Improving Step-Wise Verification via Selective Latent Steering

2026-05-20 · Yefan Zhou, Yilun Zhou, Austin Xu, Soroush Vosoughi 외 arxiv

Generative verifiers have emerged as a promising paradigm for step-wise verification, but their verification behavior is often poorly calibrated: they may be under-critical and miss erroneous steps, or over-critical and …

ConVer: Using Contracts and Loop Invariant Synthesis for Scalable Formal Software Verification

2026-05-26 · Muhammad A. A. Pirzada, Weiqi Wang, Yiannis Charalambous, Konstantin Korovin 외 arxiv

Formal verification of large C programs is impeded by state-space explosion: Bounded Model Checking (BMC) tools must encode the entire state space up to the predetermined bound by unrolling all nested constructs. We pres…

A Chain-of-Thought Is as Strong as Its Weakest Link: A Benchmark for Verifiers of Reasoning Chains

2024-02-01 · Alon Jacovi, Yonatan Bitton, Bernd Bohnet, Jonathan Herzig 외

Prompting language models to provide step-by-step answers (e.g., "Chain-of-Thought") is the prominent approach for complex reasoning tasks, where more accurate reasoning chains typically improve downstream task performan…

Open-Domain Question AnsweringQuestion Answering

Pseudo-Formalization for Automatic Proof Verification

2026-05-19 · Slim Barkallah, Luke Bailey, Kaiyue Wen, Mohammed Abouzaid 외 arxiv

Reliable verification of proofs remains a bottleneck for training and evaluating AI systems on hard mathematical reasoning. Fully formal proofs, in languages like Lean, are easy to verify because they are unambiguous and…

Mathematical Reasoning

Verifying Non-friendly Formal Verification Designs: Can We Start Earlier?

2024-10-24 · Bryan Olmos, Daniel Gerl, Aman Kumar, Djones Lettnin

The design of Systems on Chips (SoCs) is becoming more and more complex due to technological advancements. Missed bugs can cause drastic failures in safety-critical environments leading to the endangerment of lives. To o…