# t-gems: text-guided exit modules for decreasing clip image encoder Source: [t-gems: text-guided exit modules for decreasing clip image encoder]() Collection index: [llms.txt]() Data updated: 2026-05-20T05:00:13.583Z Authors: Alberto Presta, Grzegorz Stefanski, Michal Byra, Krzysztof Arendt Published: 2026-05-17T15:18:20.000Z ## Abstract Multimodal deep neural networks enhance deep comprehension by integrating diverse data modalities. Data from different modalities are typically projected into a shared latent space for similarity computation, but this process is resource intensive due to large image encoders and equal processing of test data during prediction. Early exit methods reduce computational load by utilizing intermediate layers, saving time and memory. However, developing such methods is challenging for multimodal data like image-text pairs. This study investigates the semantic content distributions present in intermediate layers of encoders such as CLIP, which can be derived from textual descriptions. We introduce Text-Guided Exit Modules (T-GEMs) and a rate-based regularizer to control encoder usage costs while maintaining cross-modal understanding performance. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Image Classification on CIFAR10 (test) (Custom Metrics RB1-RB12)](): Image Classification; CIFAR10 (test); Custom Metric RB12: 91. ## Original sources and implementation [Original paper]() [PDF]()