580 citations · 939 across the 14 of their papers we have counts for
15 papers
COMED: The Missing Middle Between Routing and Collaboration in Multi-LLM Inference
Norah Alballa, Wenxuan Zhang, Salma Kharrat +4
No single Large Language Model (LLM) is uniformly reliable across queries, motivating multi-model inference systems that either route among models or combine their outputs. However…
Scaling Agentic Reinforcement Learning for Tool-Integrated Reasoning in VLMs
Meng Lu, Ran Xu, Yi Fang +14
While recent vision-language models (VLMs) demonstrate strong image understanding, their ability to "think with images", i.e., to reason through multi-step visual interactions, rem…
Query-based Knowledge Transfer for Heterogeneous Learning Environments
Norah Alballa, Wenxuan Zhang, Ziquan Liu +3
Decentralized collaborative learning under data heterogeneity and privacy constraints has rapidly advanced. However, existing solutions like federated learning, ensembles, and tran…
Bi-Factorial Preference Optimization: Balancing Safety-Helpfulness in Language Models
Wenxuan Zhang, Philip H. S. Torr, Mohamed Elhoseiny +1
Fine-tuning large language models (LLMs) on human preferences, typically through reinforcement learning from human feedback (RLHF), has proven successful in enhancing their capabil…
Continual Learning on a Diet: Learning from Sparsely Labeled Streams Under Constrained Computation
Wenxuan Zhang, Youssef Mohamed, Bernard Ghanem +3
We propose and study a realistic Continual Learning (CL) setting where learning algorithms are granted a restricted computational budget per time step while training. We apply this…
Label Delay in Online Continual Learning
Botos Csaba, Wenxuan Zhang, Matthias Müller +4
Online continual learning, the process of training models on streaming data, has gained increasing attention in recent years. However, a critical aspect often overlooked is the lab…
Overcoming Generic Knowledge Loss with Selective Parameter Update
Wenxuan Zhang, Paul Janson, Rahaf Aljundi +1
Foundation models encompass an extensive knowledge base and offer remarkable transferability. However, this knowledge becomes outdated or insufficient over time. The challenge lies…
OxfordTVG-HIC: Can Machine Make Humorous Captions from Images?
Runjia Li, Shuyang Sun, Mohamed Elhoseiny +1
This paper presents OxfordTVG-HIC (Humorous Image Captions), a large-scale dataset for humour generation and understanding. Humour is an abstract, subjective, and context-dependent…
A Simple Baseline that Questions the Use of Pretrained-Models in Continual Learning
Paul Janson, Wenxuan Zhang, Rahaf Aljundi +1
With the success of pretraining techniques in representation learning, a number of continual learning methods based on pretrained models have been proposed. Some of these methods d…
Efficiently Disentangle Causal Representations
Yuanpeng Li, Joel Hestness, Mohamed Elhoseiny +2
This paper proposes an efficient approach to learning disentangled representations with causal mechanisms based on the difference of conditional probabilities in original and new d…
Uncertainty-guided Continual Learning with Bayesian Neural Networks
Sayna Ebrahimi, Mohamed Elhoseiny, Trevor Darrell +1
Continual learning aims to learn new tasks without forgetting previously learned ones. This is especially challenging when one cannot access data from previous tasks and when the m…
On Tiny Episodic Memories in Continual Learning
Arslan Chaudhry, Marcus Rohrbach, Mohamed Elhoseiny +4
In continual learning (CL), an agent learns from a stream of tasks leveraging prior experience to transfer knowledge to future tasks. It is an ideal framework to decrease the amoun…
Efficient Lifelong Learning with A-GEM
Arslan Chaudhry, Marc'Aurelio Ranzato, Marcus Rohrbach +1
In lifelong learning, the learner is presented with a sequence of tasks, incrementally building a data-driven prior which may be leveraged to speed up learning of a new task. In th…
Exploring the Challenges towards Lifelong Fact Learning
Mohamed Elhoseiny, Francesca Babiloni, Rahaf Aljundi +3
So far life-long learning (LLL) has been studied in relatively small-scale and relatively artificial setups. Here, we introduce a new large-scale alternative. What makes the propos…
Memory Aware Synapses: Learning what (not) to forget
Rahaf Aljundi, Francesca Babiloni, Mohamed Elhoseiny +2
Humans can learn in a continuous manner. Old rarely utilized knowledge can be overwritten by new incoming information while important, frequently used knowledge is prevented from b…