Publications

Unified Game Moderation: Soft-Prompting and LLM-Assisted Label Transfer for Resource-Efficient Toxicity Detection
Toxicity detection in gaming communities faces significant scaling challenges when expanding across multiple games and languages, particular… (voir plus)ly in real-time environments where computational efficiency is crucial. We present two key findings to address these challenges while building upon our previous work on ToxBuster, a BERT-based real-time toxicity detection system. First, we introduce a soft-prompting approach that enables a single model to effectively handle multiple games by incorporating game-context tokens, matching the performance of more complex methods like curriculum learning while offering superior scalability. Second, we develop an LLM-assisted label transfer framework using GPT-4o-mini to extend support to seven additional languages. Evaluations on real game chat data across French, German, Portuguese, and Russian achieve macro F1-scores ranging from 32.96% to 58.88%, with particularly strong performance in German, surpassing the English benchmark of 45.39%. In production, this unified approach significantly reduces computational resources and maintenance overhead compared to maintaining separate models for each game and language combination. At Ubisoft, this model successfully identifies an average of 50 players, per game, per day engaging in sanctionable behavior.
Variation Matters: from Mitigating to Embracing Zero-Shot NAS Ranking Function Variation
Pavel Rumiantsev
Mark J. Coates
Neural Architecture Search (NAS) is a powerful automatic alternative to manual design of a neural network. In the zero-shot version, a fast … (voir plus)ranking function is used to compare architectures without training them. The outputs of the ranking functions often vary significantly due to different sources of randomness, including the evaluated architecture's weights' initialization or the batch of data used for calculations. A common approach to addressing the variation is to average a ranking function output over several evaluations. We propose taking into account the variation in a different manner, by viewing the ranking function output as a random variable representing a proxy performance metric. During the search process, we strive to construct a stochastic ordering of the performance metrics to determine the best architecture. Our experiments show that the proposed stochastic ordering can effectively boost performance of a search on standard benchmark search spaces.
Visual-Tactile Inference of 2.5D Object Shape From Marker Texture.
François Robert Hogan
Charlotte Morissette
Michael Jenkin
Visual-tactile sensing affords abundant capabilities for contact-rich object manipulation tasks including grasping and placing. Here we intr… (voir plus)oduce a shape-from-texture inspired contact shape estimation approach for visual-tactile sensors equipped with visually distinct membrane markers. Under a perspective projection camera model, measurements related to the change in marker separation upon contact are used to recover surface shape. Our approach allows for shape sensing in real time, without requiring network training or complex assumptions related to lighting, sensor geometry or marker placement. Experiments show that the surface contact shape recovered is qualitatively and quantitatively consistent with those obtained through the use of photometric stereo, the current state of the art for shape recovery in visual-tactile sensors. Importantly, our approach is applicable to a large family of sensors not equipped with photometric stereo hardware, and also to those with semi-transparent membranes. The recovery of surface shape affords new capabilities to these sensors for robotic applications, such as the estimation of contact and slippage in object manipulation tasks (Hogan etal., 2022) and the use of force matching for kinesthetic teaching using multimodal visual-tactile sensing (Ablett etal., 2024).
Warmup Generations: A Task-Agnostic Approach for Guiding Sequence-to-Sequence Learning with Unsupervised Initial State Generation
Senyu Li
Jiayi Wang
Xue Liu
Pontus Stenetorp
Traditional supervised fine-tuning (SFT) strategies for sequence-to-sequence tasks often train models to directly generate the target output… (voir plus). Recent work has shown that guiding models with intermediate steps, such as keywords, outlines, or reasoning chains, can significantly improve performance, coherence, and interpretability. However, these methods often depend on predefined intermediate formats and annotated data, limiting their scalability and generalizability. In this work, we introduce a task-agnostic framework that enables models to generate intermediate "warmup" sequences. These warmup sequences, serving as an initial state for subsequent generation, are optimized to enhance the probability of generating the target sequence without relying on external supervision or human-designed structures. Drawing inspiration from reinforcement learning principles, our method iteratively refines these intermediate steps to maximize their contribution to the final output, similar to reward-driven optimization in reinforcement learning with human feedback. Experimental results across tasks such as translation, summarization, and multi-choice question answering for logical reasoning show that our approach outperforms traditional SFT methods, and offers a scalable and flexible solution for sequence-to-sequence tasks.
In Which Areas of Technical AI Safety Could Geopolitical Rivals Cooperate?
BEN BUCKNALL
SAAD SIDDIQUI
LARA THURNHERR
CONOR MCGURK
BEN HARACK
Anka Reuel
PATRICIA PASKOV
CASEY MAHONEY
Scott Singer
VINAY HIREMATH
Charbel-Raphael Segerie
OSCAR DELANEY
Alessandro Abate
Fazl Barez
Michael K. Cohen
Philip Torr
FERENC HUSZÁR
ANISOARA CALINESCU
GABRIEL DAVIS JONES … (voir 2 de plus)
Robert Trager
International cooperation is common in AI research, including between geopolitical rivals. While many experts advocate for greater internati… (voir plus)onal cooperation on AI safety to address shared global risks, some view cooperation on AI with suspicion, arguing that it can pose unacceptable risks to national security. However, the extent to which cooperation on AI safety poses such risks, as well as provides benefits, depends on the specific area of cooperation. In this paper, we consider technical factors that impact the risks of international cooperation on AI safety research, focusing on the degree to which such cooperation can advance dangerous capabilities, result in the sharing of sensitive information, or provide opportunities for harm. We begin by why nations historically cooperate on strategic technologies and analyse current US-China cooperation in AI as a case study. We further argue that existing frameworks for managing associated risks can be supplemented with consideration of key risks specific to cooperation on technical AI safety research. Through our analysis, we find that research into AI verification mechanisms and shared protocols may be suitable areas for such cooperation. Through this analysis we aim to help researchers and governments identify and mitigate the risks of international cooperation on AI safety research, so that the benefits of cooperation can be fully realised.
A novel high-dimensional model for identifying regional DNA methylation QTLs
Kaiqiong Zhao
Archer Y. Yang
Karim Oualkacha
Yixiao Zeng
Kathleen Klein
Marie Hudson
Inés Colmegna
Sasha Bernatsky
Celia M.T. Greenwood
Varying coefficient models offer the flexibility to learn the dynamic changes of regression coefficients. Despite their good interpretabilit… (voir plus)y and diverse applications, in high-dimensional settings, existing estimation methods for such models have important limitations. For example, we routinely encounter the need for variable selection when faced with a large collection of covariates with nonlinear/varying effects on outcomes, and no ideal solutions exist. One illustration of this situation could be identifying a subset of genetic variants with local influence on methylation levels in a regulatory region. To address this problem, we propose a composite sparse penalty that encourages both sparsity and smoothness for the varying coefficients. We present an efficient proximal gradient descent algorithm that scales to high-dimensional predictor spaces, providing sparse solutions for the varying coefficients. A comprehensive simulation study has been conducted to evaluate the performance of our approach in terms of estimation, prediction and selection accuracy. We show that the inclusion of smoothness control yields much better results over sparsity-only approaches. An adaptive version of the penalty offers additional performance gains. We further demonstrate the utility of our method in identifying regional mQTLs from asymptomatic samples in the CARTaGENE cohort. The methodology is implemented in the R package sparseSOMNiBUS, available on GitHub.
Sociodemographic characteristics of SARS-CoV-2 serosurveillance studies with diverse recruitment strategies, Canada, 2020 to 2023
Matthew J. Knight
Yuan Yu
Jiacheng Chen
Sheila F. O’Brien
David L. Buckeridge
Carmen Charlton
W. Alton Russell
Serological testing was a key component of severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) surveillance. Social distancing inte… (voir plus)rventions, resource limitations, and the need for timely data led to serosurveillance studies using a range of recruitment strategies, which likely influenced study representativeness. Characterizing representativeness in surveillance is crucial to identify gaps in sampling coverage and to assess health inequities. We retrospectively analyzed three pre-existing longitudinal cohorts, two convenience samples using residual blood, and one de novo probabilistic survey conducted in Canada between April 2020 – November 2023. We calculated study specimen counts by age, sex, urbanicity, race/ethnicity, and neighborhood deprivation quintiles. We derived a ‘representation ratio’ as a simple metric to assess generalizability to a target population and various sociodemographic strata. The six studies included 1,321,675 specimens. When stratifying by age group and sex, 65% of racialized minority subgroups were moderately underrepresented (representation ratio < 0.75). Representation was generally higher for older Canadians, urban neighborhoods, and neighborhoods with low material deprivation. Rural representation was highest in a study that used outpatient laboratory blood specimens. Racialized minority representation was highest in a de novo probabilistic survey cohort. While no study had adequate representation of all subgroups, less traditional recruitment strategies were more representative of some population dimensions. Understanding demographic representativeness and barriers to recruitment are important considerations when designing population health surveillance studies.
The Romantic Historicism and The Rise of the Historical Novel in the 19th Century Romanian Literature
A.R. Olteanu
Sliding ferroelectric memories and synapses based on rhombohedral-stacked bilayer MoS2
Xiuzhen Li
Biao Qin
Yaxian Wang
Yue Xi
Zhiheng Huang
Mengze Zhao
Yalin Peng
Zitao Chen
Zitian Pan
Jundong Zhu
Chenyang Cui
Rong Yang
Wei Yang
Sheng Meng
Dongxia Shi
Xuedong Bai
Can Liu
Na Li
Kaihui Liu … (voir 3 de plus)
Kai-Wen Liu
Luojun Du
Guangyu Zhang
Exploring Compound Loss Functions for Brain Tumor Segmentation
Enhancing Surgical Safety in Conflict Zones: Implementing the WHO Checklist in North Kivu
Jacques Fadhili Bake
Claude Kasereka Masumbuko
Zacharie Tsongo Kibendelwa
Background: The WHO Surgical Safety Checklist (WHO Checklist) has been shown to effectively reduce surgical complications worldwide. However… (voir plus), implementing this checklist in conflict-affected regions like North Kivu, Democratic Republic of Congo (DRC), presents unique challenges. This study investigates the utilization of the WHO Checklist in hospitals across North Kivu to identify barriers and opportunities for improvement. Methods: A cross-sectional study was conducted across 11 hospitals (5 urban and 6 rural) in North Kivu. Surveys were administered to healthcare professionals, including surgeons, anesthesiologists, and nurses, to assess their knowledge, usage, and perceptions of the WHO Checklist. Data were analyzed using SPSS version 26. Results: The response rate was 80.3%, with a majority (59.2%) from urban hospitals. The use of the WHO Checklist was inconsistent; 60.1% reported it was not utilized in the operating room. No significant differences in checklist usage were found between urban and rural hospitals (p=0.516). Training significantly correlated with the completion rate of checklist phases (p0.001) but not with overall usage (p=0.057). Furthermore, there were no significant differences regarding the need for further training to
Torque-Aware Momentum
Efficiently exploring complex loss landscapes is key to the performance of deep neural networks. While momentum-based optimizers are widely … (voir plus)used in state-of-the-art setups, classical momentum can still struggle with large, misaligned gradients, leading to oscillations. To address this, we propose Torque-Aware Momentum (TAM), which introduces a damping factor based on the angle between the new gradients and previous momentum, stabilizing the update direction during training. Empirical results show that TAM, which can be combined with both SGD and Adam, enhances exploration, handles distribution shifts more effectively, and improves generalization performance across various tasks, including image classification and large language model fine-tuning, when compared to classical momentum-based optimizers.