Nikolaos Xiros

Routing tokens through a contrastive lens boosts expert specialization and yields up to 1.77 points improvement in zero-shot reasoning accuracy.

Higher-resource languages may boost mathematical reasoning in LLMs but paradoxically reduce the faithfulness of solvability detection.

SeRIn reveals that isolating modality refinement from cross-modal integration can significantly boost sentiment analysis performance, challenging conventional fusion methods.

Verbalization, not knowledge, drives fabrication in LLMs, revealing a critical leverage point for improving mathematical reasoning accuracy.