CORTEXA
← Browse
arxivcs.LGcs.CV2026-07-12

On the modality gap and the contrastive loss in multi-modal representation learning

Fabian Mager, Hiba Nassar, Lars Kai Hansen

We study the modality gap in CLIP-style dual-encoder contrastive learning, where image and text embeddings remain misaligned despite being trained in a shared space. We argue that the gap is induced by a failure of the InfoNCE formulation with independent encoders. We conduct a uni-modal experiment with two independent encoders and identical initialization conditions and find that InfoNCE actively generates a gap at low temperatures. We provide a theoretical analysis of this phenomenon and show that the modality gap is indeed a mode-failure of InfoNCE, but only at low temperatures. We propose a simple modification called xNCE, which uses intermodal as well as intra-modality negative contrastive pairs. xNCE matches retrieval performance on MS-COCO while consistently reducing the gap even at low temperatures. Notably, xNCE improves zero-shot classification over the InfoNCE baseline across all benchmarks, whereas high-temperature InfoNCE and regularized InfoNCE both fail to do so, demonstrating that xNCE reduces the modality gap without sacrificing the discriminative geometry needed for transfer.

View free PDFSource page

Related papers

arxivcs.LGcs.CV2026-07-05

GlacierCastAI: Predicting Glacier Retreat from Multi-Modal Satellite Imagery and Climate Signals

Arunkumar Ramachandran

ERA5 seasonal climate variables contain predictive information about future glacier retreat beyond what satellite imagery alone provides, yet existing deep learning methods focus on mapping current boundaries rather than forecasting future ones. This paper presents GlacierCastAI,…

View free PDFSource page
arxivcs.LGcs.AIcs.CV2026-07-01

Multi-modal Rail Crossing Safety Analysis

Paimon Goulart, Chansong Lim, Nícolas Roque dos Santos, Yue Dong, Sheldon Peterson, Jia Chen, et al.

Given one or more images of a railway crossing, can we leverage visual cues that allow us to robustly estimate how safe it is? Can we improve our ability to do so by introducing structured data (such as official accident reports) about the accident history of that crossing into o…

View free PDFSource page
arxivcs.CVcs.LG2026-07-06

REVIVE: A Multi-Modal Framework for Vandalism Detection and Recovery in Autonomous Vehicles

Abdullah Tariq Choudhry, Tapadhir Das

Autonomous vehicles (AVs) face increasing threats from vandalism-induced occlusion attacks (VOAs) that compromise camera-based perception. While detection frameworks can identify vandalized images, restoring camera-stream utility after physical occlusion remains underexplored. Th…

View free PDFSource page
arxivcs.CVcs.LG2026-07-09

Mixture of Probes: Learning from Privileged Modalities in Multimodal LLMs Through Probing

Dominick Reilly, Qiyu Wu, Hiromi Wakaki, Srijan Das, Yuki Mistufuji

Multimodal Large Language Models (MLLMs) are typically designed under the assumption that all modalities available during training will also be accessible at inference. However, many real-world settings violate this assumption, requiring models to operate under a privileged modal…

View free PDFSource page