Publications

On the Adversarial Robustness of Discrete Image Tokenizers
Rishika Bhagwatkar
Irina Rish
Nicolas Flammarion
Francesco Croce
Discrete image tokenizers encode visual inputs as sequences of tokens from a finite vocabulary and are gaining popularity in multimodal syst… (see more)ems, including encoder-only, encoder-decoder, and decoder-only models. However, unlike CLIP encoders, their vulnerability to adversarial attacks has not been explored. Ours being the first work studying this topic, we first formulate attacks that aim to perturb the features extracted by discrete tokenizers, and thus change the extracted tokens. These attacks are computationally efficient, application-agnostic, and effective across classification, multimodal retrieval, and captioning tasks. Second, to defend against this vulnerability, inspired by recent work on robust CLIP encoders, we fine-tune popular tokenizers with unsupervised adversarial training, keeping all other components frozen. While unsupervised and task-agnostic, our approach significantly improves robustness to both unsupervised and end-to-end supervised attacks and generalizes well to unseen tasks and data. Unlike supervised adversarial training, our approach can leverage unlabeled images, making it more versatile. Overall, our work highlights the critical role of tokenizer robustness in downstream tasks and presents an important step in the development of safe multimodal foundation models.
2026-02-19
arXiv (preprint)
doi.org
arxiv.org
Dataless Weight Disentanglement in Task Arithmetic via Kronecker-Factored Approximate Curvature
Angelo Porrello
Pietro Buzzega
Felix Dangel
Thomas Sommariva
Riccardo Salami
Lorenzo Bonicelli
Simone Calderara
Task Arithmetic yields a modular, scalable way to adapt foundation models. Combining multiple task vectors, however, can lead to cross-task … (see more)interference, causing representation drift and degraded performance. Representation drift regularization provides a natural remedy to disentangle task vectors; however, existing approaches typically require external task data, conflicting with modularity and data availability constraints (e.g., privacy requirements). We propose a dataless approach by framing regularization against representation drift as a curvature matrix approximation problem. This allows us to leverage well-established techniques; in particular, we adopt Kronecker-Factored Approximate Curvature and obtain a practical regularizer that achieves state-of-the-art results in task addition and negation. Our method has constant complexity in the number of tasks and promotes robustness to task vector rescaling, eliminating the need for held-out tuning.
2026-02-18
Open MIND (preprint)
doi.org
arxiv.org
GASS: Geometry-Aware Spherical Sampling for Disentangled Diversity Enhancement in Text-to-Image Generation
Ye Zhu
Kaleb S. Newman
Johannes F. Lutzeyer
Adriana Romero
Michal Drozdzal
Olga Russakovsky
2026-02-18
ArXiv (preprint)
arxiv.org
arxiv.org
Offline-Online Retail Collaboration via Pickup Partnership
Zahra Jalali
Maxime C. Cohen
Necati Ertekin
Mehmet Gümüş
We study a growing retail strategy called pickup partnership, where online retailers partner with physical stores to offer in-store pickup s… (see more)ervices. In practice, two main policies are used in these partnerships: (i) a fixed fee policy, where the retailer pays the offline partner a set fee per pickup order, and (ii) a coupon policy, where customers receive a coupon for use at the offline partner’s store with each pickup order. Our goal is to evaluate these policies and determine which is most beneficial for online retailers. We develop a stylized model that captures the essential dynamics of pickup partnerships. We find that although the coupon policy allows the online retailer to gain greater market coverage compared with the fixed fee policy, it does not always lead to higher profits for the online retailer. The coupon policy is preferred when in-store fulfillment and pickup handling costs are low and direct-delivery costs are high, whereas the fixed fee policy is favored when these costs are moderate. We also find that both policies entail inefficiencies when the incentives of the two parties are not aligned. To alleviate such inefficiencies, we propose a new policy designed to better align incentives and improve partnership efficiency. This paper offers the first theoretical analysis of the in-store pickup partnership model and provides practical guidance for online retailers seeking to implement it. Our proposed policy aims to enhance the effectiveness and profitability of these partnerships beyond current industry practices. Supplemental Material: The online appendix is available at https://doi.org/10.1287/serv.2025.0118 .
2026-02-18
Service Science (published)
doi.org
Stackelberg Dynamic Location Planning under Cumulative Demand
Warley Almeida Silva
Margarida Carvalho
Sanjay Dominik Jena
Dynamic facility location problems predominantly suppose a monopoly over the service or product provided. Nonetheless, this premise can be a… (see more) severe oversimplification in the presence of market competitors, as customers may prefer facilities installed by one of them. The monopolistic assumption can particularly worsen planning performance when demand depends on prior location decisions of the market participants, namely, when unmet demand from one period carries over to the next. Such a demand behaviour creates an intrinsic relationship between customer demand and location decisions of all market participants, and requires the decision-maker to anticipate the competitor's response. This work studies a novel competitive facility location problem that combines cumulative demand and market competition to devise high-quality solutions. We propose bilevel mixed-integer programming formulations for two variants of our problem, prove that the optimistic variant is
2026-02-18
arXiv (preprint)
doi.org
arxiv.org
Beyond Message Passing: A Symbolic Alternative for Expressive and Interpretable Graph Learning
Chuqin Geng
Li Zhang
Haolin Ye
Ziyu Zhao
Yuhe Jiang
Tara Saba
Xinyu Wang
Xujie Si
Graph Neural Networks (GNNs) have become essential in high-stakes domains such as drug discovery, yet their black-box nature remains a signi… (see more)ficant barrier to trustworthiness. While self-explainable GNNs attempt to bridge this gap, they often rely on standard message-passing backbones that inherit fundamental limitations, including the 1-Weisfeiler-Lehman (1-WL) expressivity barrier and a lack of fine-grained interpretability. To address these challenges, we propose SymGraph, a symbolic framework designed to transcend these constraints. By replacing continuous message passing with discrete structural hashing and topological role-based aggregation, our architecture theoretically surpasses the 1-WL barrier, achieving superior expressiveness without the overhead of differentiable optimization. Extensive empirical evaluations demonstrate that SymGraph achieves state-of-the-art performance, outperforming existing self-explainable GNNs. Notably, SymGraph delivers 10x to 100x speedups in training time using only CPU execution. Furthermore, SymGraph generates rules with superior semantic granularity compared to existing rule-based methods, offering great potential for scientific discovery and explainable AI.
2026-02-17
arXiv (preprint)
doi.org
arxiv.org
Causality is Key for Interpretability Claims to Generalise
Shruti Joshi
Aaron Mueller
David Klindt
Wieland Brendel
Patrik Reizinger
Dhanya Sridhar
Interpretability research on large language models (LLMs) has yielded important insights into model behaviour, yet recurring pitfalls persis… (see more)t: findings that do not generalise, and causal interpretations that outrun the evidence. Our position is that causal inference specifies what constitutes a valid mapping from model activations to invariant high-level structures, the data or assumptions needed to achieve it, and the inferences it can support. Specifically, Pearl's causal hierarchy clarifies what an interpretability study can justify. Observations establish associations between model behaviour and internal components. Interventions (e.g., ablations or activation patching) support claims how these edits affect a behavioural metric (e.g., average change in token probabilities) over a set of prompts. However, counterfactual claims -- i.e., asking what the model output would have been for the same prompt under an unobserved intervention -- remain largely unverifiable without controlled supervision. We show how causal representation learning (CRL) operationalises this hierarchy, specifying which variables are recoverable from activations and under what assumptions. Together, these motivate a diagnostic framework that helps practitioners select methods and evaluations matching claims to evidence such that findings generalise.
2026-02-17
arXiv (preprint)
doi.org
arxiv.org
Neural Proposals, Symbolic Guarantees: Neuro-Symbolic Graph Generation with Hard Constraints
Chuqin Geng
Li Zhang
Mark Zhang
Haolin Ye
Ziyu Zhao
Xujie Si
We challenge black-box purely deep neural approaches for molecules and graph generation, which are limited in controllability and lack forma… (see more)l guarantees. We introduce Neuro-Symbolic Graph Generative Modeling (NSGGM), a neurosymbolic framework that reapproaches molecule generation as a scaffold and interaction learning task with symbolic assembly. An autoregressive neural model proposes scaffolds and refines interaction signals, and a CPU-efficient SMT solver constructs full graphs while enforcing chemical validity, structural rules, and user-specific constraints, yielding molecules that are correct by construction and interpretable control that pure neural methods cannot provide. NSGGM delivers strong performance on both unconstrained generation and constrained generation tasks, demonstrating that neuro-symbolic modeling can match state-of-the-art generative performance while offering explicit controllability and guarantees. To evaluate more nuanced controllability, we also introduce a Logical-Constraint Molecular Benchmark, designed to test strict hard-rule satisfaction in workflows that require explicit, interpretable specifications together with verifiable compliance.
2026-02-17
Open MIND (preprint)
doi.org
arxiv.org
Joint Rolling Stock and Crew Scheduling with Multi-train Composition in Urban Rail Networks
Entai Wang
Lixing Yang
Erick Delage
Jean-François Cordeau
Ziyou Gao
Yossiri Adulyasak
Rolling stock scheduling and crew scheduling are two fundamental problems that arise in the planning of urban rail operations and that are e… (see more)specially important in the case of flexible operations in real-world networks. These problems are often solved separately and sequentially in different planning stages, resulting in limited options to adjust crew schedules after rolling stock decisions have been made. To better adjust these two decision-making processes and achieve better solutions, this paper studies a joint rolling stock and crew scheduling problem in urban rail networks. A novel optimization model is formulated with the aim of reducing the operational cost of rolling stock units and crew members. In addition, the multi-train composition mode is considered to adequately match different frequency requirements and rolling stock transport capacities. To solve the model, a customized branch-and-price-and-cut solution algorithm is proposed to find the optimal schedule schemes, in which Benders decomposition is used to solve the linear programming relaxation of the path-based reformulation. Two customized column generation methods with label correcting are embedded to solve the master problem and pricing subproblem for generating paths (columns) corresponding to rolling stock units and crew groups, respectively. Finally, a branch-and-bound procedure with several acceleration techniques is proposed to find integer solutions. To demonstrate the computational performance and the robustness of the proposed approaches, a series of numerical experiments are performed in real-world instances of the Beijing urban rail network under different settings. The computational results confirm the high efficiency of the solution methodology and the benefits of the flexible operation schemes based on the solutions found by the proposed methods. Funding: This work was supported by National Natural Science Foundation of China [Grants 72288101, 72322022, 72371015]. The first author sincerely thanks the China Scholarship Council for supporting his visiting PhD program [Grant 202407090173]. Supplemental Material: The electronic companion is available at https://doi.org/10.1287/trsc.2024.0905 .
2026-02-16
Transportation Science (published)
doi.org
Continual Test-Time Adaptation: A Comprehensive Survey
Sarthak Kumar Maharana
Shambhavi Mishra
Yunbei Zhang
shuaicheng niu
Taki Hasan Rafi
Jihun Hamm
Marco Pedersoli
Jose Dolz
Guo, Yunhui
Deep neural nets achieve remarkable performance when training and test data share the same distribution, but this assumption frequently brea… (see more)ks in real-world deployment, where data undergoes continual distributional shifts. Continual Test-Time Adaptation (CTTA) addresses this challenge by adapting pretrained models to non-stationary target distributions on-the-fly, without access to source data or labeled targets, while mitigating two critical failure modes: catastrophic forgetting of source knowledge and error accumulation from noisy pseudo-labels over extended time horizons. In this comprehensive survey, we formally define the CTTA problem, analyze the diverse continual domain shift patterns that characterize different evaluation protocols, and propose a hierarchical taxonomy that categorizes existing methods into three families: optimization-based strategies (entropy minimization, pseudo-labeling, parameter restoration), parameter-efficient methods (normalization layer adaptation, adaptive parameter selection), and architecture-based approaches (teacher-student frameworks, adapters, visual prompting, masked modeling). We systematically review representative methods within each category and present comparative benchmarks and experimental results across standard evaluation settings. Finally, we discuss limitations of current approaches and highlight emerging research directions, including adaptation of foundation models and black-box systems, providing a roadmap for future research in robust continual test-time adaptation. We encourage visiting our repository at https://github.com/sarthaxxxxx/Awesome-Continual-Test-Time-Adaptation.
2026-02-15
Open MIND (preprint)
doi.org
Delta-Crosscoder: Robust Crosscoder Model Diffing in Narrow Fine-Tuning Regimes
Aly M. Kassem
Thomas Jiralerspong
Negar Rostamzadeh
Golnoosh Farnadi
Model diffing methods aim to identify how fine-tuning changes a model's internal representations. Crosscoders approach this by learning shar… (see more)ed dictionaries of interpretable latent directions between base and fine-tuned models. However, existing formulations struggle with narrow fine-tuning, where behavioral changes are localized and asymmetric. We introduce Delta-Crosscoder, which combines BatchTopK sparsity with a delta-based loss prioritizing directions that change between models, plus an implicit contrastive signal from paired activations on matched inputs. Evaluated across 10 model organisms, including synthetic false facts, emergent misalignment, subliminal learning, and taboo word guessing (Gemma, LLaMA, Qwen; 1B-9B parameters), Delta-Crosscoder reliably isolates latent directions causally responsible for fine-tuned behaviors and enables effective mitigation, outperforming SAE-based baselines, while matching the Non-SAE-based. Our results demonstrate that crosscoders remain a powerful tool for model diffing.
2026-02-15
arXiv (preprint)
doi.org
arxiv.org
Fluid-Agent Reinforcement Learning
Shishir Sharma
Doina Precup
Theodore J. Perkins
The primary focus of multi-agent reinforcement learning (MARL) has been to study interactions among a fixed number of agents embedded in an … (see more)environment. However, in the real world, the number of agents is neither fixed nor known a priori. Moreover, an agent can decide to create other agents (for example, a cell may divide, or a company may spin off a division). In this paper, we propose a framework that allows agents to create other agents; we call this a fluid-agent environment. We present game-theoretic solution concepts for fluid-agent games and empirically evaluate the performance of several MARL algorithms within this framework. Our experiments include fluid variants of established benchmarks such as Predator-Prey and Level-Based Foraging, where agents can dynamically spawn, as well as a new environment we introduce that highlights how fluidity can unlock novel solution strategies beyond those observed in fixed-population settings. We demonstrate that this framework yields agent teams that adjust their size dynamically to match environmental demands.
2026-02-15
ArXiv (preprint)
doi.org
arxiv.org