Recent advancements in MLLM-based long-form video understanding have mitigated inference-time computational cost and limited context lengths by selecting query-relevant frames. However, existing approaches predominantly rely on external proxy scorers and rigid heuristic rules, inevitably suffering from misalignment with the target MLLM's intrinsic evidence and failing to accommodate the non-uniform spatiotemporal information density.
PACE introduces a training‑free Condense‑and‑Extract framework that speeds up Vision‑Language Model inference by first adaptively downsampling visual inputs before encoding and then selectively retaining essential tokens during decoding. The Adaptive Pixel Compressor (APC) reduces encoder workload while preserving global context, and the Dynamic Dual‑Attention Extractor (DDAE) keeps task‑critical details by fusing visual and language signals. Applied to Qwen2.5‑VL‑7B, PACE maintains 93.8% of performance using only 10% of visual tokens, achieving a 3.1× speedup in time to first token.
By Junjie Liu, Shengyuan Ye, Xu Chen
arXiv:2607. 02612v1 Announce Type: cross Abstract: Vision Transformers achieve strong image classification accuracy but process all image regions with nearly the same computation, even when many regions are redundant or uninformative.
By Aravind Pradeep, Samira Nazari, Mahdi Taheri, Christian Herglotz
FLEET is a token‑based feature extractor that processes event camera data directly, using random Fourier features and cross‑attention to compress variable‑length event streams into fixed‑size latent representations. By decoupling inference cost from sensor resolution, it avoids the high compute and temporal blurring associated with CNN‑based grid aggregation. Experiments on a new high‑throughput benchmark show that FLEET outperforms state‑of‑the‑art methods and remains robust across different observation frequencies.
By Tristan Gottwald, Maximilian Schier, Melanie Schaller, Bodo Rosenhahn
The paper proposes a modular training pipeline for zero‑shot cross‑city object detection that combines a multi‑dataset pre‑training strategy with class‑agnostic objectness distillation and a domain‑resilient augmentation stream featuring a Grayworld transformation. Applied to the RF‑DETR detector, the approach reduces cross‑city distribution gaps while using only 16 GB GPU memory, achieving a 24.29‑point mAP improvement and 1st place on the AI City Challenge Track 6 leaderboard. The authors provide code and data at the referenced GitHub repository.
By Long Hoang Pham, Quoc Pham-Nam Ho, Huy-Hung Nguyen, Duong Nguyen-Ngoc Tran, Ngoc Doan-Minh Huynh, Cu Quoc Le, Hoang-Khang Nguyen, Hyung-Min Jeon, Chi Dai Tran, Son Hong Phan, Duong Khac Vu, Trinh Le Ba Khanh, Jae Wook Jeon
arXiv:2606. 01503v1 Announce Type: cross Abstract: Unified vision-language models (VLMs) integrate visual understanding and visual generation within a single autoregressive backbone, but their joint training is computationally expensive and largely overlooked from an efficiency perspective.
By Siyi Chen, Weiming Zhuang, Jingtao Li, Lingjuan Lv
Falcon Perception-HD applies reinforcement learning (GRPO) to autoregressive perception models, aligning them directly with precision and recall metrics rather than relying on maximum‑likelihood fine‑tuning. The RL framework introduces reward design for set‑structured outputs and multi‑head sampling control, enabling state‑of‑the‑art performance in very dense scenes (up to 500 objects) and eliminating common issues such as mask repetitions, NMS, and coordinate deduplication. Hybrid self‑annotation pipelines tailored for difficult referring expressions and dense scenes further boost RL training, with improvements observed across all difficulty levels on PBench and SACO‑Gold, and the model preserves object existence knowledge without negative samples.
arXiv:2604. 16557v2 Announce Type: replace Abstract: Current post-training methodologies for adapting Large Vision-Language Models (LVLMs) generally fall into two paradigms: Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL).
By Yuming Yan, Kai Tang, Sihong Chen, Ke Xu, Dan Hu, Qun Yu, Pengfei Hu
Event cameras generate asynchronous, high-frequency data streams offering spatially sparse information at lower latency than traditional cameras. In principle, these properties should be ideal for the design of control policies.
arXiv:2608. 03979v1 Announce Type: cross Abstract: We introduce Video-DeepResearch (Video-DR), extending multimodal agents from static images to continuous video streams, a setting that demands dense spatiotemporal grounding coupled with open-web exploration.
By Zhen Fang, Yu Zeng, Wenxuan Huang, Yiming Zhao, Shiting Huang, Tianfei Ren, Qi Lu, Qingnan Ren, Qisheng Su, Lionel Z. Wang, Qingyu Yin, Shuang Chen, Zehui Chen, Lin Chen, Zhenfei Yin, Yao Hu, Shaohui Lin, Wanli Ouyang, Shaosheng Cao, Feng Zhao
The paper investigates how Vision‑Language‑Action (VLA) models can generalise across different driving environments and camera setups. It introduces a multi‑dataset training strategy and an auxiliary objective called BEV‑Forcing, which injects bird‑eye‑view spatial information into the VLA backbone to improve both in‑distribution and out‑of‑distribution performance on a limited number of camera rigs. The authors observe that while BEV‑Forcing helps when training data is scarce, its advantage diminishes as the number of training embodiments grows, suggesting that scaling diversity may reduce the impact of such auxiliary tasks.
By Caio Azevedo, Stefano Sabatini, Sascha Hornauer, Fabien Moutarde
arXiv:2607. 15942v1 Announce Type: cross Abstract: Remote sensing vision-language models are increasingly expected to support open-ended reasoning over Earth Observation data and a variety of tasks.
By Stefan Maria Ailuro (INSAIT, Sofia University "St. Kliment Ohridski"), Mario Markov (INSAIT, Sofia University "St. Kliment Ohridski"), Mohammad Mahdi (INSAIT, Sofia University "St. Kliment Ohridski"), Luc Van Gool (INSAIT, Sofia University "St. Kliment Ohridski"), Danda Pani Paudel (INSAIT, Sofia University "St. Kliment Ohridski")