Tootfinder

Consistent View Alignment Improves Foundation Models for 3D Medical Image Segmentation
Puru Vaish, Felix Meister, Tobias Heimann, Christoph Brune, Jelmer M. Wolterink
https://arxiv.org/abs/2509.13846

Consistent View Alignment Improves Foundation Models for 3D Medical Image Segmentation
Many recent approaches in representation learning implicitly assume that uncorrelated views of a data point are sufficient to learn meaningful representations for various downstream tasks. In this work, we challenge this assumption and demonstrate that meaningful structure in the latent space does not emerge naturally. Instead, it must be explicitly induced. We propose a method that aligns representations from different views of the data to align complementary information without inducing false…

FRIT: Using Causal Importance to Improve Chain-of-Thought Faithfulness
Anand Swaroop, Akshat Nallani, Saksham Uboweja, Adiliia Uzdenova, Michael Nguyen, Kevin Zhu, Sunishchal Dev, Ashwinee Panda, Vasu Sharma, Maheep Chaudhary
https://arxiv.org/abs/2509.13334

FRIT: Using Causal Importance to Improve Chain-of-Thought Faithfulness
Chain-of-thought (CoT) reasoning has emerged as a powerful tool for improving large language model performance on complex tasks, but recent work shows that reasoning steps often fail to causally influence the final answer, creating brittle and untrustworthy outputs. Prior approaches focus primarily on measuring faithfulness, while methods for systematically improving it remain limited. We introduce Faithful Reasoning via Intervention Training (FRIT), a scalable alignment method that trains mode…

Do LLMs Align Human Values Regarding Social Biases? Judging and Explaining Social Biases with LLMs
Yang Liu, Chenhui Chu
https://arxiv.org/abs/2509.13869 https://

Do LLMs Align Human Values Regarding Social Biases? Judging and Explaining Social Biases with LLMs
Large language models (LLMs) can lead to undesired consequences when misaligned with human values, especially in scenarios involving complex and sensitive social biases. Previous studies have revealed the misalignment of LLMs with human values using expert-designed or agent-based emulated bias scenarios. However, it remains unclear whether the alignment of LLMs with human values differs across different types of scenarios (e.g., scenarios containing negative vs. non-negative questions). In this…

DuetUI: A Bidirectional Context Loop for Human-Agent Co-Generation of Task-Oriented Interfaces
Yuan Xu, Shaowen Xiang, Yizhi Song, Ruoting Sun, Xin Tong
https://arxiv.org/abs/2509.13444

DuetUI: A Bidirectional Context Loop for Human-Agent Co-Generation of Task-Oriented Interfaces
Large Language Models are reshaping task automation, yet remain limited in complex, multi-step real-world tasks that require aligning with vague user intent and enabling dynamic user override. From a formative study with 12 participants, we found that end-users actively seek to shape generative interfaces rather than relying on one-shot outputs. To address this, we introduce the human-agent co-generation paradigm, materialized in DuetUI. This LLM-empowered system unfolds alongside task progress…

On the quadratic complexity of subsets of $\mathbb{F}_p^n$ of bounded $\mathrm{VC_{2}}$-dimension
C. Terry, J. Wolf
https://arxiv.org/abs/2510.12767 https://

On the quadratic complexity of subsets of $\mathbb{F}_p^n$ of bounded $\mathrm{VC_{2}}$-dimension
In prior work, we showed that subsets of $\mathbb{F}_{p}^{n}$ of $\mathrm{VC_{2}}$-dimension at most $k$ are well approximated by a union of atoms of a quadratic factor of complexity $(\ell,q)$, where the complexity $\ell$ of the linear part and the complexity $q$ of the quadratic part are both bounded in terms of $k$, $p$, and the desired level of approximation $μ$. A key tool in the proof of this result was an arithmetic regularity lemma for the Gowers $U^3$-norm by Green and Tao, which resu…

Machines are more productive than humans until they aren't, and vice versa
Riccardo Zanardelli
https://arxiv.org/abs/2509.14057 https://arxiv.org/pdf/2…

Machines are more productive than humans until they aren't, and vice versa
With the growth of artificial skills, organizations may increasingly confront with the problem of optimizing skill policy decisions guided by economic principles. This paper addresses the underlying complexity of this challenge by developing an in-silico framework based on Monte Carlo simulations grounded in empirical realism to analyze the economic impact of human and machine skills, individually or jointly deployed, in the execution of tasks presenting varying levels of complexity. Our result…

ViCO: A Training Strategy towards Semantic Aware Dynamic High-Resolution
Long Cui, Weiyun Wang, Jie Shao, Zichen Wen, Gen Luo, Linfeng Zhang, Yanting Zhang, Yu Qiao, Wenhai Wang
https://arxiv.org/abs/2510.12793

ViCO: A Training Strategy towards Semantic Aware Dynamic High-Resolution
Existing Multimodal Large Language Models (MLLMs) suffer from increased inference costs due to the additional vision tokens introduced by image inputs. In this work, we propose Visual Consistency Learning (ViCO), a novel training algorithm that enables the model to represent images of varying semantic complexities using different numbers of vision tokens. The key idea behind our method is to employ multiple MLP connectors, each with a different image compression ratio, to downsample the vision …

Hey Dashboard!: Supporting Voice, Text, and Pointing Modalities in Dashboard Onboarding
Vaishali Dhanoa, Gabriela Molina Le\'on, Eve Hoggan, Eduard Gr\"oller, Marc Streit, Niklas Elmqvist
https://arxiv.org/abs/2510.12386

Hey Dashboard!: Supporting Voice, Text, and Pointing Modalities in Dashboard Onboarding
Visualization dashboards are regularly used for data exploration and analysis, but their complex interactions and interlinked views often require time-consuming onboarding sessions from dashboard authors. Preparing these onboarding materials is labor-intensive and requires manual updates when dashboards change. Recent advances in multimodal interaction powered by large language models (LLMs) provide ways to support self-guided onboarding. We present DIANA (Dashboard Interactive Assistant for Na…

VQArt-Bench: A semantically rich VQA Benchmark for Art and Cultural Heritage
A. Alfarano (University of Zurich, Max Planck Society), L. Venturoli (University of Zurich, Max Planck Society), D. Negueruela del Castillo (University of Zurich, Max Planck Society)
https://arxiv.org/abs/2510.12750

VQArt-Bench: A semantically rich VQA Benchmark for Art and Cultural Heritage
Multimodal Large Language Models (MLLMs) have demonstrated significant capabilities in joint visual and linguistic tasks. However, existing Visual Question Answering (VQA) benchmarks often fail to evaluate deep semantic understanding, particularly in complex domains like visual art analysis. Confined to simple syntactic structures and surface-level attributes, these questions fail to capture the diversity and depth of human visual inquiry. This limitation incentivizes models to exploit statisti…

Task-Specific Dual-Model Framework for Comprehensive Traffic Safety Video Description and Analysis
Blessing Agyei Kyem, Neema Jakisa Owor, Andrews Danyo, Joshua Kofi Asamoah, Eugene Denteh, Tanner Muturi, Anthony Dontoh, Yaw Adu-Gyamfi, Armstrong Aboah
https://arxiv.org/abs/2510.11907

Task-Specific Dual-Model Framework for Comprehensive Traffic Safety Video Description and Analysis
Traffic safety analysis requires complex video understanding to capture fine-grained behavioral patterns and generate comprehensive descriptions for accident prevention. In this work, we present a unique dual-model framework that strategically utilizes the complementary strengths of VideoLLaMA and Qwen2.5-VL through task-specific optimization to address this issue. The core insight behind our approach is that separating training for captioning and visual question answering (VQA) tasks minimizes…