AH
Ahmed Heakl
cs.AIcs.LGcs.CLcs.CVcs.CYcs.HCcs.MMcs.ROcs.ARcs.NE
On Valency
published · living versionsW_sc3d9k3q·v1 · currentpublished
MedRoundsQA: A Persona and Difficulty Aware Evaluation for Multi-Turn Medical Consultations
with Youssef Mohamed, Qinrong Cui, Junhong Liang, Rafiq Ali +11
1 version
Preprints & journals
30 papers in the corpus · 2022–2026Program-Verified Self-Evolution for Vision-Language Models2609.33855v1 · Ahmed Heakl, Sungik Choi, Moontae Lee et al.2026 · 0 citationsarXiv
MedRoundsQA: A Persona and Difficulty Aware Evaluation for Multi-Turn Medical Consultations2609.12851v1 · Youssef Mohamed, Ahmed Heakl, Qinrong Cui et al.2026 · 0 citationsarXivon Valency
Right Frame, Wrong Rule: Cultural Cues Expose the Financial Knowledge Gap They Were Meant to Close2609.00999v1 · Rania Elbadry, Ahmed Heakl, Saeed Almheiri et al.2026 · 0 citationsarXiv
CEPO: RLVR Self-Distillation using Contrastive Evidence Policy Optimization2605.19436v1 · Ahmed Heakl, Abdelrahman M. Shaker, Youssef Mohamed et al.2026 · 0 citationsarXiv
DocAtlas: Multilingual Document Understanding Across 80+ Languages2605.12623v2 · Ahmed Heakl, Youssef Mohamed, Abdullah Sohail et al.2026 · 0 citationsarXiv
The Geometry of Forgetting: Temporal Knowledge Drift as an Independent Axis in LLM Representations2605.09195v1 · Rania Elbadry, Ahmed Heakl, Fan Zhang et al.2026 · 0 citationsarXiv
SAHM: A Benchmark for Arabic Financial and Shari'ah-Compliant Reasoning2604.19098v2 · Rania Elbadry, Sarfraz Ahmad, Ahmed Heakl et al.2026 · 1 citationACL 2026
WorldCache: Content-Aware Caching for Accelerated Video World Models2603.22286v1 · Umair Nawaz, Ahmed Heakl, Ufaq Khan et al.2026 · 0 citationsarXiv
Latent-DARM: Bridging Discrete Diffusion And Autoregressive Models For Reasoning2603.09184v1 · Lina Berrayana, Ahmed Heakl, Abdullah Sohail et al.2026 · 0 citationsarXiv
MASEval: Extending Multi-Agent Evaluation from Models to Systems2603.08835v1 · Cornelius Emde, Alexander Rubinstein, Anmol Goel et al.2026 · 0 citationsarXiv
Mobile-O: Unified Multimodal Understanding and Generation on Mobile Device2602.20161v2 · Abdelrahman Shaker, Ahmed Heakl, Jaseel Muhammad et al.2026 · 0 citationsarXiv
Planner and Executor: Collaboration between Discrete Diffusion And Autoregressive Models in Reasoning2510.15244v2 · Lina Berrayana, Ahmed Heakl, Muhammad Abdullah Sohail et al.2025 · 0 citationsarXiv
Dr.LLM: Dynamic Layer Routing in LLMs2510.12773v2 · Ahmed Heakl, Martin Gubri, Salman Khan et al.2025 · 0 citationsarXiv
How Good are Foundation Models in Step-by-Step Embodied Reasoning?2509.15293v2 · Dinura Dissanayake, Ahmed Heakl, Omkar Thawakar et al.2025 · 0 citationsarXiv
VideoMolmo: Spatio-Temporal Grounding Meets Pointing2506.05336v2 · Ghazi Shazan Ahmad, Ahmed Heakl, Hanan Gani et al.2025 · 0 citationsarXiv
SVRPBench: A Realistic Benchmark for Stochastic Vehicle Routing Problem2505.21887v2 · Ahmed Heakl, Yahia Salaheldin Shaaban, Martin Takac et al.2025 · 1 citationarXiv
CASS: Nvidia to AMD Transpilation with Data, Models, and Benchmark2505.16968v4 · Ahmed Heakl, Gustavo Bertolo Stahl, Sarim Hashmi et al.2025 · 1 citationACL 2026
KITAB-Bench: A Comprehensive Multi-Domain Benchmark for Arabic OCR and Document Understanding2502.14949v2 · Ahmed Heakl, Abdullah Sohail, Mukul Ranjan et al.2025 · 14 citationsarXiv
AIN: The Arabic INclusive Large Multimodal Model2502.00094v2 · Ahmed Heakl, Sara Ghaboura, Omkar Thawkar et al.2025 · 0 citationsarXiv
LlamaV-o1: Rethinking Step-by-step Visual Reasoning in LLMs2501.06186v1 · Omkar Thawakar, Dinura Dissanayake, Ketan More et al.2025 · 19 citationsarXiv
From CISC to RISC: language-model guided assembly transpilation2411.16341v1 · Ahmed Heakl, Chaimaa Abi, Rania Hossam et al.2024 · 0 citationsarXiv
CAMEL-Bench: A Comprehensive Arabic LMM Benchmark2410.18976v1 · Sara Ghaboura, Ahmed Heakl, Omkar Thawakar et al.2024 · 2 citationsarXiv
Precision Aquaculture: An Integrated Computer Vision and IoT Approach for Optimized Tilapia Feeding2409.08695v3 · Rania Hossam, Ahmed Heakl, Walid Gomaa2024 · 14 citationsarXiv
ArzEn-LLM: Code-Switched Egyptian Arabic-English Translation and Speech Recognition Using LLMs2406.18120v2 · Ahmed Heakl, Youssef Zaghloul, Mennatullah Ali et al.2024 · 9 citationsarXiv
ResumeAtlas: Revisiting Resume Classification with Large-Scale Datasets and Large Language Models2406.18125v2 · Ahmed Heakl, Youssef Mohamed, Noran Mohamed et al.2024 · 27 citationsarXiv
Arabic Handwritten Text for Person Biometric Identification: A Deep Learning Approach2406.00409v1 · Mazen Balat, Youssef Mohamed, Ahmed Heakl et al.2024 · 3 citationsarXiv
DroneVis: Versatile Computer Vision Library for Drones2406.00447v1 · Ahmed Heakl, Fatma Youssef, Victor Parque et al.2024 · 1 citationarXiv
Advancing Ear Biometrics: Enhancing Accuracy and Robustness through Deep Learning2406.00135v1 · Youssef Mohamed, Zeyad Youssef, Ahmed Heakl et al.2024 · 5 citationsarXiv
AraSpider: Democratizing Arabic-to-SQL2402.07448v1 · Ahmed Heakl, Youssef Mohamed, Ahmed B. Zaky2024 · 1 citationarXiv
A Study on Broadcast Networks for Music Genre Classification2208.12086v1 · Ahmed Heakl, Abdelrahman Abdelgawad, Victor Parque2022 · 15 citationsarXiv
Career total: 38 works. 30 are in this corpus.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.