Visual Jev: Accurate and Efficient Decisions from Shared Visual Context

arXiv:2609.25845 · cs.CV, cs.LG · Submitted 2026-09-22 · Read on arXiv

cs.CV, cs.LG

Submitted: 2026-09-22

Updated: 2026-09-22

Comments: Code: https://github.com/guanxuyu-sv/Visual-Jev

Code: https://github.com/guanxuyu-sv/Visual-Jev

License: http://creativecommons.org/licenses/by-nc-nd/4.0/

The gist: Many vision applications ask several independent, forced-choice questions about the same image.

Terminology

Abstract

Many vision applications ask several independent, forced-choice questions about the same image. Visual Jev encodes the image and public context once, executes isolated question suffixes as a batch, and reads candidate probabilities from the backbone's language-model head. Across four benchmarks, answer-supervised post-training raises equal-weight macro accuracy from 70.6% to 76.1%, with the gain concentrated on the two task families represented in training. At N=32 questions per image, shared batched execution is 8.9x faster in warm amortized time than independent serial execution and remains 3.4x faster than an already-batched baseline that recomputes the prefix, at the cost of higher peak memory. A matched typed-head control offers no consistent accuracy advantage over the language-model-head readout. The supported design is therefore simple: adapt the backbone for quality, retain the existing readout, and share execution for efficiency.

Sources

Related papers