Papers for

advertising analytics teams

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Multimodal model reduces gender bias in economic meme interpretation

Addressing Overcommitment in the Reasoning of Gendered Economic Memes under Multimodal Ambiguity

Abstract: Multimodal meme understanding is increasingly used to analyze socially sensitive content, yet existing models often exhibit biased behavior when interpreting economic dependence and social roles under ambiguity. Many memes express economic relationships through sparse text or symbolic visual cues, providing insufficient evidence for gendered attribution. In such underspecified settings, models tend to rely on pretraining correlations, leading to hallucinated and stereotypical economic role assignments. In this work, we study gendered economic dependence in image-text memes through the lens of contextual sufficiency and identify epistemic overcommitment-inferring roles without adequate evidence-as a primary source of bias. We propose CGER-Net, a context-grounded multimodal framework that estimates whether the input provides sufficient evidence for gendered economic reasoning and applies evidence-gated inference to enable confident attribution when cues are explicit while favoring principled abstention otherwise. We evaluate CGER-Net on EconMeme-GE, a curated dataset of image-text memes annotated as Men, Women, Neutral, or Ambiguous. Across strong contemporary multimodal baselines, CGER-Net reduces Gender Overcommitment Rate by up to 44% on ambiguous instances while maintaining comparable accuracy on unambiguous cases. Human evaluation further shows that 79% of generated rationales are judged as epistemically aligned with the available evidence. These results highlight the importance of modeling when not to infer for reliable and responsible multimodal analysis.

Thu 8 OctMachine Learning
The gist
Images and text combined in memes can be confusing, especially when they hint at gender roles in money matters but don't give enough clues. The authors found that many computer models make guesses based on stereotypes when there isn’t clear evidence, which causes bias. They designed a new system called CGER-Net that checks whether enough information exists before guessing the gender roles and can choose to say 'I don’t know' if it’s unclear. This approach led to fewer wrong gender guesses while still being accurate when evidence was clear.
Open → 2610.11724v1

CRISP method cuts training data drastically while preserving fraud detection accuracy

CRISP: Scalable Importance-Stratified Coresets for Imbalanced Tabular Learning

Abstract: Large imbalanced tabular datasets make repeated gradient-boosted tree training expensive. Existing coreset methods often lose accuracy when most majority examples are removed. We present CRISP (Coreset Reduction via Importance-Stratified Pruning), a linear-time method that allocates a negative-class budget across quantile strata of a proxy-model score. Sample weights account for unequal inclusion probabilities. At 95% negative-class reduction on a production fraud dataset, CRISP trains on approximately 1.70M of 25M rows and retains 99.7% of full-data Average Precision. This is a 93.2% reduction in total training rows. On public CriteoPrivateAds, CRISP has the highest mean Average Precision at each tested rate from 90% to 99.4% majority reduction. Sparkov results are mixed at lower rates, but CRISP has the highest mean at 99.2% and 99.4%. Ablations identify budget allocation and inverse-propensity weighting as the main sources of the production-dataset gain.

Tue 22 SeptMachine Learning
The gist
Large datasets with many more normal examples than rare ones make training machine learning models slow and expensive. The authors present CRISP, a new way to smartly pick a much smaller subset of data while keeping most of the model’s accuracy. This method keeps very few normal examples but chooses them carefully based on how important they seem, allowing faster training with almost the same results. They tested CRISP on real fraud and advertising data, showing big savings in training time without losing performance.
Open → 2609.26962v1