Tootfinder

Enhancing Concept Localization in CLIP-based Concept Bottleneck Models
R\'emi Kazmierczak, Steve Azzolin, Elo\"ise Berthier, Goran Frehse, Gianni Franchi
https://arxiv.org/abs/2510.07115

Enhancing Concept Localization in CLIP-based Concept Bottleneck Models
This paper addresses explainable AI (XAI) through the lens of Concept Bottleneck Models (CBMs) that do not require explicit concept annotations, relying instead on concepts extracted using CLIP in a zero-shot manner. We show that CLIP, which is central in these techniques, is prone to concept hallucination, incorrectly predicting the presence or absence of concepts within an image in scenarios used in numerous CBMs, hence undermining the faithfulness of explanations. To mitigate this issue, we …

En noticias futbolísticas que importan: Deportes Concepción vuelve a primera división de Chile después de vencer a Cobreloa en la final de la liguilla ppr el ascenso.
https://chile.as.com/futbol/concepcion

Post-hoc Stochastic Concept Bottleneck Models
Wiktor Jan Hoffmann, Sonia Laguna, Moritz Vandenhirtz, Emanuele Palumbo, Julia E. Vogt
https://arxiv.org/abs/2510.08219 https://

Post-hoc Stochastic Concept Bottleneck Models
Concept Bottleneck Models (CBMs) are interpretable models that predict the target variable through high-level human-understandable concepts, allowing users to intervene on mispredicted concepts to adjust the final output. While recent work has shown that modeling dependencies between concepts can improve CBM performance, especially under interventions, such approaches typically require retraining the entire model, which may be infeasible when access to the original data or compute is limited. I…

📸 Honor has a wild idea for getting steady smartphone shots
#camera

Honor has a wild idea for getting steady smartphone shots
Imagine a smartphone powered by AI with a gimbal for the camera. Honor's latest smartphone concept could become the first robot in your pocket.

#AdventOfSystemSeeing Day 5, still trying to catch up with @…

Advent(ure) in System Seeing — Day 5 — Key Systems Concept: Feedback
Let’s gather some of the thoughts that have been percolating around systems. Let’s assume that you are designing the introductory page of…

Trump Loyalists Push 'Grand Conspiracy' as New Subpoenas Land (New York Times)
https://www.nytimes.com/2025/11/09/us/politics/trump-conspiracy-probe-subpoenas.html
http://www.memeorandum.com/251109/p24#a251109p24

Sources: UEFA's next soccer broadcast rights sale could fetch £4.3B per year, up from £2.9B per year in the 2024 deal, as it targets streaming platforms (BBC)
https://www.bbc.com/sport/football/articles/c5y8njv2ylvo

European clubs as Uefa looks to streaming services
The top European clubs could cash in on streaming rights as Uefa look to land around £4bn in their next TV deal.

The J6 Pipe Bomber Arrest Only Made the Conspiracy Theories Worse (Will Sommer/The Bulwark)
https://www.thebulwark.com/p/j6-pipe-bomber-arrest-only-made-conspiracy-theories-worse-maga
http://www.memeorandum.com/251208/p146#a251208p146

Concept Retrieval -- What and How?
Ori nizan, Oren Shrout, Ayellet Tal
https://arxiv.org/abs/2510.07058 https://arxiv.org/pdf/2510.07058

Concept Retrieval - What and How?
A concept may reflect either a concrete or abstract idea. Given an input image, this paper seeks to retrieve other images that share its central concepts, capturing aspects of the underlying narrative. This goes beyond conventional retrieval or clustering methods, which emphasize visual or semantic similarity. We formally define the problem, outline key requirements, and introduce appropriate evaluation metrics. We propose a novel approach grounded in two key observations: (1) While each neighb…

A Modular Algorithm for Non-Stationary Online Convex-Concave Optimization
Qing-xin Meng, Xia Lei, Jian-wei Liu
https://arxiv.org/abs/2509.07901 https://arx…

A Modular Algorithm for Non-Stationary Online Convex-Concave Optimization
This paper investigates the problem of Online Convex-Concave Optimization, which extends Online Convex Optimization to two-player time-varying convex-concave games. The goal is to minimize the dynamic duality gap (D-DGap), a critical performance measure that evaluates players' strategies against arbitrary comparator sequences. Existing algorithms fail to deliver optimal performance, particularly in stationary or predictable environments. To address this, we propose a novel modular algorithm wit…