analiza aiqbench
Omanic introduces a novel 4-hop QA benchmark that moves beyond final-answer accuracy to diagnose specific breakdowns in LLM reasoning chains, such as error propagation and knowledge floors. By providing structured graph topologies and step-level annotations, this dataset offers a rigorous tool for evaluating and improving the intermediate logic capabilities of large language models.
Omanic: Towards Step-wise Evaluation of Multi-hop Reasoning in Large Language Models
arXiv:2603.16654v3 Announce Type: replace-cross Abstract: Evaluating the reasoning abilities of large language models (LLMs) solely from final answers can obscure failures in intermediate steps, especially in multi-hop QA benchmarks without step-level annotations. To address this gap, we introduce Omanic, an open-domain 4-hop QA benchmark designed not only to measure final-answer accuracy but also to diagnose where reasoning breaks down. Omanic contains 10,296 machine-generated training examples (OmanicSynth) and 967 expert-reviewed human-annotated evaluation examples (OmanicBench), with each evaluation question decomposed into single-hop sub-questions, intermediate answers, and structured graph topologies. Experiments with proprietary and open-source LLMs show that Omanic is challenging, while step-wise analysis reveals a later-hop bottleneck, factual knowledge floor, and error propagation along reasoning chains. Fine-tuning on OmanicSynth transfers to six reasoning and mathematics benchmarks, yielding a 7.41-point average gain and validating its effectiveness as supervision for reasoning-capability transfer. We release the data at https://huggingface.co/datasets/li-lab/Omanic and the code at https://github.com/XiaojieGu/Omanic.
Authors:Xiaojie Gu, Sherry T. Tong, Aosong Feng, Sophia Simeng Han, Jinghui Lu, Yingjian Chen, Yusuke Iwasawa, Yutaka Matsuo, Chanjun Park, Rex Ying, Irene Li
Abstract:Evaluating the reasoning abilities of large language models (LLMs) solely from final answers can obscure failures in intermediate steps, especially in multi-hop QA benchmarks without step-level annotations. To address this gap, we introduce Omanic, an open-domain 4-hop QA benchmark designed not only to measure final-answer accuracy but also to diagnose where reasoning breaks down. Omanic contains 10,296 machine-generated training examples (OmanicSynth) and 967 expert-reviewed human-annotated evaluation examples (OmanicBench), with each evaluation question decomposed into single-hop sub-questions, intermediate answers, and structured graph topologies. Experiments with proprietary and open-source LLMs show that Omanic is challenging, while step-wise analysis reveals a later-hop bottleneck, factual knowledge floor, and error propagation along reasoning chains. Fine-tuning on OmanicSynth transfers to six reasoning and mathematics benchmarks, yielding a 7.41-point average gain and validating its effectiveness as supervision for reasoning-capability transfer. We release the data at this https URL and the code at this https URL.
Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
arXiv:2603.16654 [cs.CL]
(or arXiv:2603.16654v3 [cs.CL] for this version)
Text integral sub licența deschisă a publicației · 190 cuvinte
Subiecte conexe
J++ Lens: Jacobian Filtering Enables More Faithful Workspace Lenses
LessWrong · 8 oct. 2026
State of the Field: AI for Epistemics and Coordination
LessWrong · 2 oct. 2026
oMeBench: Towards Robust Benchmarking of LLMs in Organic Mechanism Elucidation and Reasoning
arXiv · 18 sept. 2026
The Continual Constitutional Convention: How language models revise the documents that govern them
AI Alignment Forum · 13 oct. 2026