arXiv Machine Learning
MERIT: Mitigating Exposure Bias in Generative XMC for User-Interest Propensity Modeling
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
arXiv Machine Learning
Jul 17
Deep-learning Causal Retrieval Optimization for Efficient e-commerce Distribution in Pinterest
arXiv:2607. 14161v1 Announce Type: cross Abstract: Pinterest is where people turn inspiration into action as users browse ideas, then take steps toward realization, often by discovering shoppable content.
arXiv AI
Jul 7
LBR: Towards Mitigating Length Bias in Large Language Models for Recommendation
arXiv:2607. 04270v1 Announce Type: cross Abstract: Large language models (LLMs) have recently emerged as powerful backbones for recommender systems by reformulating recommendation as a token-level generation task.
Hugging Face Trending Papers
Jul 5
LBR: Towards Mitigating Length Bias in Large Language Models for Recommendation
Large language models (LLMs) have recently emerged as powerful backbones for recommender systems by reformulating recommendation as a token-level generation task. Despite their promise, we identify a pervasive yet underexplored issue: $\textit{Length Bias}$.
arXiv Machine Learning
Jun 17
On the Memorization Behavior of LLMs in Generative Recommendation: Observations, Implications, and Training Strategies
arXiv:2606. 17276v1 Announce Type: cross Abstract: Generative recommendation (GR) has emerged as a promising direction for recommender systems.
arXiv AI
Jul 31
Improving Item Discoverability in e-Commerce Search via Related Intent Generation
arXiv:2607. 27172v1 Announce Type: cross Abstract: Traditional search systems are optimized to retrieve items that strictly match a query, often prioritizing precision over recall.
Hugging Face Trending Papers
Aug 10
UNMASK: Discovering and Causally Verifying Spurious Shortcuts in Text Classifiers
Neural language models trained on large crowdsourced corpora frequently exploit spurious surface patterns tied to target labels without true linguistic or causal relevance, boosting benchmark performance while failing on adversarial or out-of-distribution inputs. Existing approaches either require manual specification of the feature vocabulary or automate discovery only partially, leaving the gap between dataset-level correlation and model-level exploitation unaddressed.