MAMahmud Elahi Akhter

Mahmud Elahi Akhter

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Mahmud Elahi Akhter has a total of 23 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    The Alignment Bottleneck in Decomposition-Based Claim Verification

    by Mahmud Elahi Akhter, Federico Ruggeri, Iman Munire Bilal, Rob Procter, Maria Liakata

    Structured claim decomposition is often proposed as a solution for verifying complex, multi-faceted claims, yet empirical results have been inconsistent. We argue that these inconsistencies stem from two overlooked bottlenecks: evidence alignment and sub-claim error profiles. To better understand these factors, we introduce a new dataset of real-world complex claims, featuring temporally bounded evidence and human-annotated sub-claim evidence spans. We evaluate decomposition under two evidence alignment setups: Sub-claim Aligned Evidence (SAE) and Repeated Claim-level Evidence (SRE). Our results reveal that decomposition brings significant performance improvement only when evidence is granular and strictly aligned. By contrast, standard setups that rely on repeated claim-level evidence (SRE) fail to improve and often degrade performance as shown across different datasets and domains (PHEMEPlus, MMM-Fact, COVID-Fact). Furthermore, we demonstrate that in the presence of noisy sub-claim labels, the nature of the error ends up determining downstream robustness. We find that conservative "abstention" significantly reduces error propagation compared to aggressive but incorrect predictions. These findings suggest that future claim decomposition frameworks must prioritize precise evidence synthesis and calibrate the label bias of sub-claim verification models.

  • Open Access

    Fundamental Reasoning Paradigms Induce Out-of-Domain Generalization in Language Models

    by Mingzi Cao, Xingwei Tan, Mahmud Elahi Akhter, Marco Valentino, Maria Liakata, Xi Wang, Nikolaos Aletras

    Deduction, induction, and abduction are fundamental reasoning paradigms, core for human logical thinking. Although improving Large Language Model (LLM) reasoning has attracted significant research efforts, the extent to which the fundamental paradigms induce generalization has yet to be systematically explored. In this study, we shed light on how the interplay between these core paradigms influences LLMs' reasoning behavior. To this end, we first collect a new dataset of reasoning trajectories from symbolic tasks, each targeting one of the three fundamental paradigms, to abstract from concrete world knowledge. Then, we investigate effective ways for inducing these skills into LLMs. We experiment with a battery of methods including simple fine-tuning, and more complex approaches to increase model depth, or transform a dense model to a mixture-of-experts. We comprehensively evaluate induced models on realistic out-of-domain tasks, that are entirely formulated in natural language and contain real-world knowledge. Our results reveal that our approach yields strong generalizability with substantial performance gains (up to $14.60$) across realistic tasks.

  • A Novel Generative Inverse Approach Towards Silicon-Based Nano-Photonic Power Splitter Design Generation

    by Mahmud Elahi Akhter, Ibraheem Muhammad Moosa, Lubaba Tazrian Rahman, Mohammad Atiqul Islam, Sharnali Islam, Khaleda Ali

    In recent years, the design of chip-based photonic systems has significantly moved towards Artificial Intelligence-assisted data-driven methods instead of conventional intuition and simulation-based ones. The time-consuming nature of traditional chip-design methods, coupled with their insufficient flexibility to accommodate rapidly evolving integrated circuit requirements, contributes to this situation. In this work, we propose a novel generative model for the inverse design of nanophotonic power splitters. Our proposed model generates power splitters from arbitrary response spectra from 1.46 to 1.63 μm with a central wavelength of 1.55 μm. The model employs machine learning and a quadratic programming solver, which consists of a linear regressor and a mixed integer quadric programming solver. It is deterministic due to its generator being a quadratic programming solver. We empirically show that the generated structures have error margins within 10-4% and 2×10-4% for any given arbitrary response spectra. Furthermore, we also show that the model is capable of handling and generating Out-of-Distribution responses and their associated devices.


  • Open Access

    Assessing the Reasoning Capabilities of LLMs in the context of Evidence-based Claim Verification

    by John Dougrez-Lewis, Mahmud Elahi Akhter, Federico Ruggeri, Sebastian Löbbers, Yulan He, Maria Liakata

    John Dougrez-Lewis, Mahmud Elahi Akhter, Federico Ruggeri, Sebastian Löbbers, Yulan He, Maria Liakata. Findings of the Association for Computational Linguistics: ACL 2025. 2025.


  • COVID-19 Diagnosis and Classification from CXR Images Using Vision Transformer

    by Mahbubur Rahman, Shihabur Rahman Samrat, Abdullah Al Ahad, Mahmud Elahi Akhter, Ibraheem Muhammad Moosa, Rajesh Palit, Ashfia Binte Habib