aiqbench analysis
Bactrainus addresses the critical evidence bottleneck in multi-hop question answering by introducing a modular framework that separates paragraph selection from answer generation to mitigate vulnerability to irrelevant context. The study demonstrates through ablation experiments on Llama 3.1 models that model scale alone cannot resolve sensitivity to distractors, highlighting instead the necessity of explicit supervision and sentence-level control for robust performance. These findings offer actionable insights into optimizing large language models for complex reasoning tasks where integrating distributed evidence is essential.
Bactrainus: Optimizing Large Language Models for Multi-hop Complex Question Answering Tasks
arXiv:2501.06286v2 Announce Type: replace-cross Abstract: Multi-hop question answering requires a system to identify and integrate evidence distributed across documents, yet large language models remain vulnerable to irrelevant context. We investigate this evidence bottleneck in the English HotpotQA distractor setting and introduce Bactrainus, a modular selector-reader framework that separates paragraph selection, supporting-sentence identification, and answer generation. Optional question decomposition and teacher-generated rationale supervision make it possible to test where additional reasoning structure is useful. The evaluation combines foundation-model screening, controlled context and prompting ablations, parameter-efficient adaptation of Llama 3.1 8B Instruct and Llama 3.1 70B Instruct readers, and integrated selector-reader experiments. Supplying the full candidate context instead of gold supporting facts reduces answer token-overlap F1 by 17-21 points, showing that scale alone does not remove context sensitivity. The largest observed differences are associated with reader adaptation and sentence-level evidence control. The strongest reported configuration obtains 89.01 answer F1 and 79.70 joint F1, whereas decomposition and rationale-supervision variants yield smaller, recipe-dependent changes. These findings support auditable, explicitly supervised evidence interfaces for fixed-candidate multi-hop QA and motivate blind, matched, multi-seed evaluation of the remaining small differences.
Abstract:Multi-hop question answering requires a system to identify and integrate evidence distributed across documents, yet large language models remain vulnerable to irrelevant context. We investigate this evidence bottleneck in the English HotpotQA distractor setting and introduce Bactrainus, a modular selector-reader framework that separates paragraph selection, supporting-sentence identification, and answer generation. Optional question decomposition and teacher-generated rationale supervision make it possible to test where additional reasoning structure is useful. The evaluation combines foundation-model screening, controlled context and prompting ablations, parameter-efficient adaptation of Llama 3.1 8B Instruct and Llama 3.1 70B Instruct readers, and integrated selector-reader experiments. Supplying the full candidate context instead of gold supporting facts reduces answer token-overlap F1 by 17-21 points, showing that scale alone does not remove context sensitivity. The largest observed differences are associated with reader adaptation and sentence-level evidence control. The strongest reported configuration obtains 89.01 answer F1 and 79.70 joint F1, whereas decomposition and rationale-supervision variants yield smaller, recipe-dependent changes. These findings support auditable, explicitly supervised evidence interfaces for fixed-candidate multi-hop QA and motivate blind, matched, multi-seed evaluation of the remaining small differences.
Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
arXiv:2501.06286 [cs.CL]
(or arXiv:2501.06286v2 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2501.06286 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 274 words
Related coverage
[AINews] not much happened today
Latent Space · Sep 18, 2026
[AINews] High Taste Testers Ep. 1
Latent Space · Oct 17, 2026
J++ Lens: Jacobian Filtering Enables More Faithful Workspace Lenses
LessWrong · Oct 8, 2026
Presentation: From AI Agent Demo to Production: Automated Testing and Evaluation
InfoQ · Sep 7, 2026