The paper introduces Interventional Transfer (IT), a method for evaluating rubric generation by assessing how two rubrics respond to perturbations that cause a pass/fail shift. IT reveals asymmetries in rubrics generated by different LLMs when applied to HealthBench, showing that improvements on one rubric do not reliably transfer to the other. The study contrasts IT with existing evaluation approaches, highlighting its ability to uncover these asymmetries.
By Erik Skalnes, Layne C. Price, Raviteja Anantha, Michael Oberst
The paper introduces RuVerBench, a benchmark with 2,458 instances for evaluating the reliability of Large Language Models acting as judges (LaaJ) in verifying rubric compliance within agentic scenarios such as deep research and agentic coding. It systematically meta‑evaluates frontier LLMs, revealing that even the most advanced models perform well yet still produce substantial noise. The study also examines how prompt design, batching, and majority voting affect verification accuracy, noting that weaker models are more prompt‑sensitive, batched verification trades accuracy for efficiency, and majority voting offers diminishing returns.
By Yangda Peng, Yunjia Qi, Haotian Xia, Guanzhong He, Xintong Shi, Richeng Xuan, Songyuanyi Lu, Yixian Liu, Zhichao Hu, Yuhong Liu, Hao Peng
arXiv:2603. 05659v3 Announce Type: replace-cross Abstract: Reinforcement learning with verifiable rewards (RLVR) and Rubrics as Rewards (RaR) have driven strong gains in domains with clear correctness signals and even in subjective domains by synthesizing evaluation criteria from ideal reference answers.
By Wisdom Ikezogwo, Mehmet Saygin Seyfioglu, Ranjay Krishna, Karim Bouyarmane
arXiv:2610.17757v1 Announce Type: new
Abstract: Large language model agents can follow complex human instructions to solve real-world tasks. In practice, however, initial instructions are often incom...
By Kaiming Liu, Yuxuan Liu, Jinrui Ju, Minyuan Ruan, Xuanyu Lei, Peng Li, Yang Liu
Interlock is a framework that transforms long‑horizon tasks into typed completion graphs, linking evidence, decisions, and required effects through grounding, coupling, and realization edges. It compiles these graphs into a resettable workspace, a public request, and an evaluation rubric, and uses source‑coverage, public‑input, and agent‑as‑judge reviews to ensure that all mandatory requirements are met. Fine‑tuning Qwen3.6 backbones on Interlock‑generated trajectories improves pass rates on AutomationBench, JobBench, and Harvey LAB, showing that the synthesized experience benefits external task environments.
By Linji Hao, Yanji He, Zihan Wang, Hansi Yang, Jiayuan Zhang, Hongyan Hao, Qi Gu, Yan Feng, Han Wu
Law students are examined through open-text answers: case solutions, essays and memos that apply the law to a set of facts. Such answers are expensive to grade, human graders disagree widely on them,...
CriticGen introduces a generation‑aware evaluation framework that generates sample‑specific evaluation dimensions and scoring criteria across categories such as subjective, objective, and self‑derived constraints. These dynamic rubrics produce a score, reason, executable refinement suggestion, and a refined answer, enabling models to diagnose and target flaws in their responses. Experiments show significant gains in rubric quality, score correlation, and actionable feedback, with 73.17% of answers improved and a 93.28% non‑degradation rate.
By Huifang Du, Zecheng Zuo, Sen Wang, Chenghao Fan, Haofen Wang, Yehui Yang
The paper introduces ImpossibleRubrics, a benchmark of 169 impossible tasks designed to test the robustness of language‑model‑generated rubrics as reward signals. Each task is paired with a verifiable oracle certificate that defines what constitutes an honest answer, and the benchmark includes 48 answerable controls. Experiments show that many rubric generators are exploited frequently—up to 36% on a stress cut—highlighting a significant gap in rubric quality rather than task difficulty, and that generic rubrics can be more vulnerable than tailored ones.
By Bowen Qin, Yi Xie, Yesheng Liu, Xi Yang
arXiv:2607. 25675v1 Announce Type: new Abstract: Text-space optimization adapts large language models (LLMs) by editing external natural-language artifacts rather than model weights, so the optimized artifacts remain inspectable and the model can be treated as a black box.
By Jiangwang Chen, Zixin Song, Junlin Liu, Shuaiyu Zhou, Haiyan Wu, Haihan Shi, Chenxi Zhou, Hanqing Li, Xiao Yang, Da Zhu, Guanjun Jiang, Hai Wan, Xibin Zhao
arXiv:2610.14918v1 Announce Type: new
Abstract: Evaluating an agent's ability to strictly satisfy diverse and procedurally complex requirements is essential to determining its practical reliability....
By Zhengxi Li, Junyi Zhou, Qiyuan Zhang, Chuhan Wu, Fuyuan Lyu, Le Tian, Xiao Zhou, Chen Ma, Yufei Wang
arXiv:2608.29856v1 Announce Type: new
Abstract: Large language models are increasingly used as scalable evaluators for open-ended tasks. However, many LLM judges derive query-specific criteria during...
By Yifan Chen, Haitao Li, Qingyao Ai, Fengbin Zhu, Tat-Seng Chua, Min Zhang, Yiqun Liu
TrussTutor-Bench is a benchmark of 200 verified two-dimensional truss problems with automatically generated student solutions, designed to evaluate large language models on solution accuracy, sycophancy, and error detection. Thirteen LLMs were tested, showing accuracy ranging from 64.5% to 97.0%, near-perfect error detection (90.4%–100%), and varying degrees of sycophancy (1.6%–50.0% of correct solutions rejected when a student doubted them). The benchmark highlights wide performance differences among models, especially in handling student doubt, and offers a verified, extensible standard for assessing technical reliability before deploying LLMs in educational settings.
By Evangelos Apostolakis, Erfan Hamdi, Sara Jahedazad, Emma Lejeune