AIConfPaper
← Search

Mennatallah El-Assady

10 accepted papers

2026

Learning Reward Functions from Multiple Feedback Types with Amortized Variational Inference

Raphaël Baur, Yannick Metz, Maria Gkoulta, Mennatallah El-Assady, Giorgia Ramponi, Thomas Kleine Buening
ICML 2026poster

Reward learning typically relies on a single feedback type or combines multiple feedback types using manually weighted loss terms. Currently, it remains unclear how to jointly learn reward functions from heterogeneous feedback types such as demonstrations, comparisons, ratings, rankings, and stops t…

Cited by 0Source ↗Scholar ↗
2025

Reward Learning from Multiple Feedback Types

Yannick Metz, Andras Geiszl, Raphaël Baur, Mennatallah El-Assady
ICLR 2025poster

Learning rewards from preference feedback has become an important tool in the alignment of agentic models. Preference-based feedback, often implemented as a binary comparison between multiple completions, is an established method to acquire large-scale human feedback. However, human feedback in othe…

Cited by 1Source ↗PDF ↗code ↗Scholar ↗
2024

Navigating the Maze of Explainable AI: A Systematic Approach to Evaluating Methods and Metrics

Lukas Klein, Carsten T. Lüth, Udo Schlegel, Till J. Bungert, Mennatallah El-Assady, Paul F Jaeger
NeurIPS 2024poster

Explainable AI (XAI) is a rapidly growing domain with a myriad of proposed methods as well as metrics aiming to evaluate their efficacy. However, current studies are often of limited scope, examining only a handful of XAI methods and ignoring underlying design parameters for performance, such as the…

Cited by 3Source ↗PDF ↗code ↗Scholar ↗
2024

On Affine Homotopy between Language Encoders

Robin Chan, Reda Boumasmoud, Anej Svete, Yuxin Ren, Qipeng Guo, Zhijing Jin +5
NeurIPS 2024poster

Pre-trained language encoders---functions that represent text as vectors---are an integral component of many NLP tasks. We tackle a natural question in language encoder analysis: What does it mean for two encoders to be similar? We contend that a faithful measure of similarity needs to be \e…

Cited by 0Source ↗PDF ↗Scholar ↗
2024

PowerGraph: A power grid benchmark dataset for graph neural networks

Anna Varbella, Kenza Amara, Blazhe Gjorgiev, Mennatallah El-Assady, Giovanni Sansavini
NeurIPS 2024poster

Power grids are critical infrastructures of paramount importance to modern society and, therefore, engineered to operate under diverse conditions and failures. The ongoing energy transition poses new challenges for the decision-makers and system operators. Therefore, we must develop grid analysis al…

Cited by 6Source ↗PDF ↗Scholar ↗
2024

SyntaxShap: Syntax-aware Explainability Method for Text Generation

Kenza Amara, Rita Sevastjanova, Mennatallah El-Assady
ACL 2024findings

To harness the power of large language models in safety-critical domains, we need to ensure the explainability of their predictions. However, despite the significant attention to model interpretability, there remains an unexplored domain in explaining sequence-to-sequence tasks using methods tailore…

Cited by 7Source ↗PDF ↗code ↗Scholar ↗
2023

A Diachronic Perspective on User Trust in AI under Uncertainty

Shehzaad Zuzar Dhuliawala, Vilém Zouhar, Mennatallah El-Assady, Mrinmaya Sachan
EMNLP 2023long main

In human-AI collaboration, users typically form a mental model of the AI system, which captures the user's beliefs about when the system performs well and when it does not. The construction of this mental model is guided by both the system's veracity as well as the system output presented to the use…

Cited by 0Source ↗code ↗Scholar ↗
2022

Automatic Generation of Socratic Subquestions for Teaching Math Word Problems

Kumar Shridhar, Jakub Macina, Mennatallah El-Assady, Tanmay Sinha, Manu Kapur, Mrinmaya Sachan
EMNLP 2022main

Socratic questioning is an educational method that allows students to discover answers to complex problems by asking them a series of thoughtful questions. Generation of didactically sound questions is challenging, requiring understanding of the reasoning process involved in the problem. We hypothes…

Cited by 49Source ↗PDF ↗code ↗Scholar ↗
2021

Explaining Contextualization in Language Models using Visual Analytics

Rita Sevastjanova, Aikaterini-Lida Kalouli, Christin Beck, Hanna Schäfer, Mennatallah El-Assady
ACL 2021long

Despite the success of contextualized language models on various NLP tasks, it is still unclear what these models really learn. In this paper, we contribute to the current efforts of explaining such models by exploring the continuum between function and content words with respect to contextualizatio…

Cited by 25Source ↗PDF ↗Scholar ↗
2020

XplaiNLI: Explainable Natural Language Inference through Visual Analytics

Aikaterini-Lida Kalouli, Rita Sevastjanova, Valeria de Paiva, Richard Crouch, Mennatallah El-Assady
COLING 2020system demonstrations

Advances in Natural Language Inference (NLI) have helped us understand what state-of-the-art models really learn and what their generalization power is. Recent research has revealed some heuristics and biases of these models. However, to date, there is no systematic effort to capitalize on those ins…

Cited by 4Source ↗PDF ↗code ↗Scholar ↗