mlsanityData to 14 Oct 2026

Visual Jev: Accurate and Efficient Decisions from Shared Visual Context

arXiv 2609.25845 ↗First posted 22 Sep 2026cs.CV cs.LGcited by 9

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. Cited by 9 papers in the map (data through 14 Oct 2026). Builds on 32 of its 45 references that are in the map. First posted on arXiv 22 Sep 2026.

Builds on

32 of 45 references.

References · 32

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,599
  3. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
  4. 2017On Calibration of Modern Neural Networkscited by 4,960
  5. 2021RoFormer: Enhanced Transformer with Rotary Position Embeddingcited by 4,703
  6. 2022Flamingo: a Visual Language Model for Few-Shot Learningcited by 4,386
  7. 2024Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolutioncited by 4,234
  8. 2021Finetuned Language Models Are Zero-Shot Learnerscited by 3,780
  9. 2022FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awarenesscited by 3,673
  10. 2023InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuningcited by 2,876
  11. 2016Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotationscited by 2,656
  12. 2016Making the V in VQA Matter: Elevating the Role of Image Understanding in Visual Question Answeringcited by 2,516
20 more
  1. 2023Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyondcited by 2,224
  2. 2022Language Models (Mostly) Know What They Knowcited by 1,856
  3. 2019Towards VQA Models That Can Readcited by 1,718
  4. 2018Know What You Don't Know: Unanswerable Questions for SQuADcited by 1,554
  5. 2023GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpointscited by 1,346
  6. 2017Mixed Precision Trainingcited by 1,156
  7. 2023SGLang: Efficient Execution of Structured Language Model Programscited by 1,141
  8. 2017Selective Classification for Deep Neural Networkscited by 852
  9. 2019Fast Transformer Decoding: One Write-Head is All You Needcited by 717
  10. 2018Annotation Artifacts in Natural Language Inference Datacited by 671
  11. 2023Large Language Models Are Not Robust Multiple Choice Selectorscited by 486
  12. 2017Don't Just Assume; Look and Answer: Overcoming Priors for Visual Question Answeringcited by 339
  13. 2018Hypothesis Only Baselines in Natural Language Inferencecited by 295
  14. 2023Large Language Models Sensitivity to The Order of Options in Multiple-Choice Questionscited by 263
  15. 2019Visual Entailment: A Novel Task for Fine-Grained Image Understandingcited by 252
  16. 2022Leveraging Large Language Models for Multiple Choice Question Answeringcited by 203
  17. 2018TallyQA: Answering Complex Counting Questionscited by 153
  18. 2022Reliable Visual Question Answering: Abstain Rather Than Answer Incorrectlycited by 89
  19. 2024DeepStack: Deeply Stacking Visual Tokens is Surprisingly Simple and Effective for LMMscited by 75
  20. 2020e-SNLI-VE: Corrected Visual-Textual Entailment with Natural Language Explanationscited by 31

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 9.

9 papers.

  1. 2026Typed Decision Models: An Early Evidence Audit and Evaluation Checklistcited by 3
  2. 2026From Text Decisions to Pixels: An Study of Jev-Style Visual Choice Modelcited by 2
  3. 2026Benchmarking System One decision models against trained classifiers and language models for automated decision gatescited by 1
  4. 2026AnyJev Technical Reportcited by 1
  5. 2026OmniMed-Jev: Calibrating LVLM Confidence for Trustworthy Medical Multimodal Decisions via System Onecited by 0
  6. 2026Visual Jev Rewards: Reference-Bound Verification for Multi-Subject Image Generationcited by 0
  7. 2026Can Jev be Your Q or Policy in Reinforcement Learning?cited by 0
  8. 2026VideoJev: System One Model for Video Understanding and Interactioncited by 0
  9. 2026DecisionRobust: Benchmarking Robustness for JEV Style Decision Models Across Text and Visioncited by 0