arXiv Computer Vision By Mengzhao Wang, Yanli Ji, Peng Ye

Lookahead Prediction-Guided Adaptive Visual Refocusing for Interleaved Multimodal Reasoning

Read the original on arXiv Computer Vision →

The paper introduces Lookahead Prediction‑Guided Adaptive Visual Refocusing (LPAVR), a test‑time framework that uses Multi‑Token Prediction (MTP) to anticipate future reasoning steps and adjust visual evidence accordingly. LPAVR compares lookahead‑guided visual evidence with history‑guided evidence, switching to the former when the latter fails to cover the needed regions, thereby aligning visual context with upcoming reasoning. Experiments on multimodal reasoning benchmarks demonstrate that LPAVR improves reasoning accuracy and increases the acceptance rates of speculative tokens, enhancing both visual reasoning and decoding efficiency.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computer Vision.

arXiv Computer Vision
Oct 9

Look Back, Think Ahead: Visual Memory on Demand for Efficient Multimodal Reasoning

The paper introduces ViMoD, a lightweight framework that dynamically manages visual memory for multimodal large language models. ViMoD uses Deformable Aggregation of Region-wise Tokens (DART) to create content‑adaptive coarse representations linked to recoverable fine tokens, and Temporal Routing for Adaptive Contextual Evidence (TRACE) to anticipate and adjust evidence needs during reasoning. On the Qwen3‑VL‑4B model, ViMoD achieves superior performance across eight reasoning benchmarks with only a 20% visual token budget and minimal additional parameters.

By Yicheng Xue, Han Wu, Jufeng Yang, Minjing Dong, Xinghao Chen, Hanting Chen, Jianyuan Guo
arXiv AI
Oct 28

VideoLVR: Latent Visual Reasoning over Grounded Video Evidence

VideoLVR introduces a method for multimodal large language models to maintain visual grounding during chain-of-thought reasoning by constructing latent visual evidence from local video regions at relevant moments. It uses a visual token prediction objective that unifies visual evidence reconstruction and text generation, allowing models to interleave text tokens with latent visual states. The approach is further extended with reinforcement learning, treating visual tokens as actions and optimizing policies with final-answer rewards, thereby improving video understanding without expanding the multimodal context.

By Shilong Li, Xingyuan Bu, Hao Shi, Yun Xing, Weixin Feng, Qiangpeng Yang, Shilei Wen
arXiv Computer Vision
Aug 25

FOVEA: Focused On-Demand Visual Evidence Adaptation for Cache-Friendly Multimodal Speculative Decoding

FOVEA introduces a cache‑friendly, on‑demand visual evidence adaptation for multimodal speculative decoding, enabling a lightweight draft model to dynamically retrieve a bounded subset of visual memory based on a cumulative‑mass rule. The retrieved visual readout is fused with the draft hidden state via a lightweight gated residual correction, avoiding the insertion of visual tokens into the autoregressive context. Experiments on various vision‑language backbones and benchmarks show that FOVEA improves draft acceptance and speeds up end‑to‑end decoding by up to 2.13× compared to traditional autoregressive decoding.

By Hengjie Zhu, Dayan Wu, Zihao Zhang, Xinze Liu, Jingxuan Yu, Peng Fu, Zheng Lin, Weiping Wang, Ding Wang
arXiv AI
5d ago

Seeing Isn't Enough: Unlocking Visual Reasoning by Bridging the Reasoning-Access Gap with TIGER

The paper introduces TIGER, a training framework that addresses the modality reasoning‑access gap in multimodal large language models (MLLMs). By generating synthetic counterfactuals where models succeed on text but fail on rendered images, TIGER fine‑tunes MLLMs using Group Relative Policy Optimization, SFT, and DPO, improving performance on real‑world visual benchmarks such as MathVerse, EMMA, MMMU, ChartQAPro, and RoboSpatial‑Home. The approach demonstrates that visual reasoning can be enhanced without multimodal annotations, and it complements existing RLVR methods.

By Gregory Kang Ruey Lau, Minh Huynh Nguyen, Bryan Kian Hsiang Low