Past ArXiv / July 22, 2026

Daily archive

A preserved paper digest from the previous month.

Archive date 07/22/2026

Source ArXiv

Back to current digest

Reading Queue

ArXiv

24 papers

Vision-Language Reasoning

1 IGGT4D: Streaming 4D Instance-Grounded Geometry Transformer Zhengyu Zou, Hao Li, Kuixuan Jiao, Liu Liu, Tingyang Xiao, Xiaolin Zhou, Fangzhou Hong, Zhizhong Su, Dingwen Zhang, Ziwei Liu

Vision-Language Reasoning Benchmark & Evaluation Embodied Interaction LLM Interpretability

ArXiv

Paper 1 / arXiv:2607.19228 Open arXiv

Relevance 9 Novelty 8

Why selected: Matches criterion 1 and criterion 3 very closely: streaming 4D instance-grounded geometry for online scene understanding, with a new dataset for dynamic spatial intelligence.

Real-world spatial intelligence requires agents to understand scenes from continuous video streams, where objects move, persist, disappear, and reappear over time. While recent spatial foundation models have enabled generalizable feed-forward 3D reconstruction, most streaming methods remain geometry-centric and lack temporally consistent object-level understanding. Meanwhile, existing semantic reconstruction and 3D-aware vision-language methods largely rely on externally extracted 2D semantic cues or loosely coupled geometry inputs, limiting unified geometry-instance learning in long dynamic scenes. In this paper, we propose IGGT4D, a streaming instance-grounded geometry Transformer for online 4D scene understanding. IGGT4D processes video frames sequentially, reuses historical context through causal spatial-temporal modeling, and incrementally updates a unified representation of camera motion, geometry, and object identity. This enables long-sequence feed-forward reconstruction with geometry-instance consistency in dynamic environments. To address the lack of high-quality 4D supervision, we further construct InsScene4D-147K, a large-scale dataset spanning real/synthetic and static/dynamic scenes, with RGB images, depth, poses, and temporally consistent instance masks generated by an automated geometry-guided annotation pipeline. Experiments on 3D reconstruction, pose estimation, instance spatial tracking, and open-vocabulary segmentation demonstrate that IGGT4D outperforms existing streaming baselines while maintaining scalable online inference for long dynamic sequences.

2 ExpertVerse: A General-Purpose Benchmark for Expert-Level Reasoning in Knowledge-Intensive Visual Synthesis Yuan Wang, Yongchao Du, Mengting Chen, Jinsong Lan, Xuetao Feng, Xiaoyong Zhu

Vision-Language Reasoning Multimodal LLM Benchmark & Evaluation Generative Modeling

ArXiv

Paper 2 / arXiv:2607.19341 Open arXiv

Relevance 8 Novelty 7

Why selected: Matches criterion 2 very closely: it is a new benchmark plus method for multimodal/visual generative models (KnowThinker, BPPO) in knowledge-intensive visual synthesis.

Recent advances in multimodal generative models have enabled instruction-based image generation to move beyond semantic manipulation to knowledge-driven visual reasoning. However, these methods focus on explicit commonsense reasoning, shallow causal understanding, and direct knowledge recall, failing at knowledge-intensive generation. We develop \textbf{ExpertVerse}, a capability-centric benchmark to evaluate generative models via knowledge-intensive lens. ExpertVerse stratifies reasoning generation across an orthogonal taxonomy of \textit{9 cognitive capabilities} and \textit{8 expert disciplines}, yielding \textit{58 sub-disciplines}. We curate 1,611 expert-annotated instances covering single-image editing, multi-image composition, and text-to-image generation. We further develop an automated workflow to produce \textbf{ExpertVerse-100K}, a large-scale dataset with reasoning traces and knowledge-anchored rationale annotations. Based on this, we train \textbf{KnowThinker} with RL fine-tuning, a VLM reasoning engine with world knowledge that jointly generates thinking processes and refined instructions. Towards the cross-modal credit misalignment and multi-objective gradient conflicts in multi-reward optimization, we propose a tailored Bootstrapped Pareto Policy Optimization (BPPO), which synergizes Bootstrapping Reward Rectification (BRR) and Conflict-Aware Pareto Advantage Fusion (CPAF). Extensive results of both open-source and proprietary models exposes critical reasoning deficits, highlighting imperative for knowledge-intensive benchmarks towards next-generation visual generation.

4 ABot-World-0: Infinite Interactive World Rollout on a Single Desktop GPU Fan Jiang, Zhaoxu Sun, Mengchao Wang, Ziyu Zhu, Chiyu Wang, Yunpeng Zhang, Wenlin Liu, Yun Wang, Xue Zheng, Rui Sun, Junfeng Ni, Hongyu Pan, Zhongxu Sun, Fei Yu, Zengye Ge, Mengmeng Du, Nianfei Fan, Mingchao Sun, Yu Liu, Yongchang, Yanqing Zhu, Jiahang Wang, Ning Ying, Yuze Xuan, Di Yang, Zhicheng Liu, Zhe Gao, Tingbing Xu, Jiacheng Sui, Wenjin Yang, Junnan Lai, Shufeng Liu, Yuan Liu, Zheng Zhou, Yingliang Peng, Dawei Cao, Kaifeng Sheng, Yuxiang Cai, Fei Lu, Mu Xu, Ning Guo

Vision-Language Reasoning Embodied Interaction LLM Interpretability

ArXiv

Paper 4 / arXiv:2607.19191 Open arXiv

Relevance 6 Novelty 7

Why selected: Strong match for criterion 3: it builds an interactive world-rollout system/world model for embodied interaction, including simulator/game/video data and long-horizon closed-loop rollout.

We present ABot-World-0, an action-conditioned video world model for real-time, long-horizon closed-loop interaction, supported by a multi-source data infrastructure spanning AAA games, simulation engines, and internet videos to learn controllable world dynamics. WorldExplorer performs agent-driven collection guided by training feedback, while a unified pipeline applies 14 deterministic quality checks, VLM-based assessment, and synchronized action and text annotation. We progressively distill a bidirectional action-conditioned teacher into a causal student through teacher forcing and ODE distillation, and introduce LongForcing to align long student self-rollouts with an extended-horizon teacher, mitigating accumulated distribution shift and autoregressive drift. Raw keyboard actions provide a unified control interface for scene roaming and third-person character interaction, while reference-character memory provides persistent appearance cues for identity consistency during third-person rollouts. For deployment, we co-design a streaming inference stack with a lightweight VAE decoder, efficient attention, memory-aware scheduling, and low-bit DiT inference. Across optimized low-bit configurations, ABot-World-0 streams 720P video at up to 16 FPS on a single NVIDIA RTX 5090 desktop GPU, with 1.2s action-to-first-frame latency and approximately 19GiB peak VRAM. Experiments on WorldRoamBench and extended interactive rollouts demonstrate competitive controllability and coherent long-horizon world evolution.

5 Attributes Should Come from Images, Not Class Names: Distribution-Conditioned Attribute Selection for Vision-Language Models Gautam Rajendrakumar Gare, Jia Shi, Zhiqiu Lin, Deepak Pathak, John Galeotti, Deva Ramanan

Vision-Language Reasoning Benchmark & Evaluation Medical AI Retrieval & Grounding

ArXiv

Paper 5 / arXiv:2607.18695 Open arXiv

Relevance 6 Novelty 7

Why selected: Matches criterion 4 closely: a vision foundation model/CLIP-style application paper with a clever statistical insight about selecting attributes from images instead of class names.

A popular route to interpretable zero-shot classification asks a large language model (LLM) to describe each class name and prompts CLIP with the resulting descriptors. We show that these descriptors carry little visual evidence of their own: removing the class name from the prompt collapses ImageNet accuracy from 59.5% to 15.5%. The diagnosis is that the descriptors are conditioned on the label rather than on the images, so they describe the concept in general and mislead exactly when the data shifts; an LLM insists that strawberries are red, but every strawberry in ImageNet-Sketch is a colorless line drawing. We therefore select attributes from the target image collection instead: we score a large attribute pool against the images in CLIP's joint embedding space and keep the top-scoring attributes per class. Selected this way, class-name-free attribute prompts reach 23.8% on ImageNet (against 15.5% for LLM descriptors), the gain holds on four shifted ImageNet variants, and reselecting from the LLM's own pool isolates the selection mechanism as the cause. With one image per class, the selected attributes outperform the prompt-tuning method CoOp by 3 points while fitting in under a minute instead of 14 hours, with no learned soft prompt to obscure the decision. Because the attribute set is chosen by the data, it doubles as a readable summary of a dataset, which we use to describe distribution shift in words.

7 DeforM: Reasoning-Guided Physics-Aware Video Generation via Spatial-Temporal Masking Yunyi Li, Yu Qiao, Yaohui Wang, Xinyuan Chen

Vision-Language Reasoning Generative Modeling LLM Interpretability Time-Series LLM

ArXiv

Paper 7 / arXiv:2607.18664 Open arXiv

Relevance 5 Novelty 7

Why selected: Strong match for criterion 4: it combines VLM-guided reasoning with physics-aware video generation, which is directly relevant to vision foundation models and generative modeling.

Video generation models achieve high visual quality but often struggle to generate physics-aware videos. Unlike rigid-body motion, which can be described by explicit trajectories or formulas, complex deformation dynamics remain challenging to synthesize. We observe that a lack of physical reasoning for localizing dynamic areas allows irrelevant regions to dilute the model's attention, leading to generation failure. In this paper, we propose DeforM, a reasoning-guided image-to-video generation framework that directs the model's focus toward physics-critical regions. To reason about and localize these critical regions, we introduce a VLM-guided physical reasoning module, DeforM-Reason, to identify target objects and generate spatial-temporal masks. For physical guidance, we develop two alternative strategies: DeforM-Free for training-free mechanism analysis and DeforM-Injection as a powerful training-based generator. Experimental results demonstrate that DeforM improves the realism of generated deformation scenarios, outperforming baseline models in both visual quality and physical consistency.

9 FilmWorld: Agentic Novel-to-Film Generation through Dynamic Cinematic World Modeling Jialong Zuo, Haotong Zuo, Shiwei Zhang, Xiang Wang, Chen Li, Nong Sang, Changxin Gao, Xiang Bai

Vision-Language Reasoning Benchmark & Evaluation Embodied Interaction Generative Modeling

ArXiv

Paper 9 / arXiv:2607.19038 Open arXiv

Relevance 5 Novelty 7

Why selected: Matches criterion 3 closely: embodied-style agentic system for long-form novel-to-film generation plus a new benchmark/evaluation suite (FilmEval) with novel consistency/fidelity angles.

Translating novels into films poses a grand challenge for generative artificial intelligence, requiring conversion of abstract literary prose into long-form, multi-scene visual narratives. While current video generation models excel at short, single-scene clips within narrow temporal and spatial contexts, novel-to-film generation operates in a more complex regime, demanding long-duration content across diverse scenes with dynamically evolving entity states. To address this, we formalize novel-to-film generation as dynamic cinematic world modeling, decomposed into two phases: construction, which grounds abstract, underspecified literary narratives into concrete, stateful, and persistent world entities; and evolution, which governs how these entities dynamically update under plot progression to maintain causal consistency across scenes. We propose FilmWorld, an end-to-end agentic system where two groups of specialized agents collaborate to instantiate these phases. Construction-side agents perform narrative structured translation, world entity state modeling with visual anchoring, and state-driven shot planning, progressively projecting literary language into a cinematic blueprint. Evolution-side agents perform state-anchored visual generation, cross-shot dynamic state propagation, and closed-loop state verification to maintain causal consistency and visual coherence. To address the evaluation gap in long-form generation, we introduce FilmEval, a systematic evaluation framework that couples a difficulty-graded benchmark of 15 representative novels with an automated protocol of nine objective metrics spanning three dimensions: cinematic presentation, film consistency, and novel fidelity. Experiments demonstrate that FilmWorld consistently outperforms state-of-the-art video generation agent systems, with particularly pronounced improvements in narrative fidelity and cross-scene consistency.

11 PathAgentBench: Benchmarking Evidence-Seeking Vision-Language Models on Whole-Slide Pathology Image Dankai Liao, Tianyi Zhang, Yufeng Wu, Xinyue Zhang, Qiaochu Xue, Zeyu Liu, Dachun Zhao, Linghan Cai, Yueming Jin

Vision-Language Reasoning Benchmark & Evaluation Embodied Interaction Medical AI

ArXiv

Paper 11 / arXiv:2607.19261 Open arXiv

Relevance 6 Novelty 6

Why selected: Matches criterion 3 closely: a new embodied-style benchmark for evidence-seeking VLMs on whole-slide pathology images, with strong emphasis on localization and multi-scale evidence acquisition.

Whole-slide image (WSI) diagnosis requires identifying diagnostically relevant regions, examining them across magnifications, and integrating multi-scale evidence. However, most existing pathology benchmarks evaluate models on pre-cropped patches or pre-extracted slide features, leaving their ability to acquire evidence directly from gigapixel WSIs largely untested. We introduce PathAgentBench, a benchmark for evaluating evidence-seeking vision-language models (VLMs) across four complementary capabilities: image-to-text matching for evidence interpretation, text-to-image retrieval for evidence verification, diagnostic-region localization for evidence acquisition, and multi-scale reasoning for evidence integration. The benchmark is organized as a diagnostic tree that links nested regions across magnifications with scale-specific findings and path-level diagnoses. It contains 1,822 TCGA WSIs and 17,135 diagnostic paths annotated by ten board-certified pathologists. An additional private cohort of 190 breast cancer WSIs with detailed annotations is used to evaluate autonomous whole-slide exploration. We evaluate 20 general-purpose, medical, and pathology-specialized models. Leading open-weight models achieve over 93% accuracy in multi-scale reasoning and over 50% accuracy in both cross-modal matching tasks. In contrast, diagnostic-region localization remains challenging: the best text-guided mean intersection-over-union is below 0.09, underperforming a simple center-based heuristic. During autonomous exploration, the unconditional hit rate decreases from 0.522 at low magnification to 0.185 at intermediate magnification and 0.020 at high magnification. These results reveal a pronounced gap between reasoning over curated evidence and acquiring that evidence directly from WSIs. PathAgentBench provides a unified framework for measuring and improving evidence-seeking pathology models.

12 Cross-Modal UAV Object Tracking: State-Aware Representation Learning and A Unified Benchmark Yun Xiao, Zhihong Hong, Jiandong Jin, Chenglong Li, Jin Tang, Amir Hussain

Vision-Language Reasoning Benchmark & Evaluation

ArXiv

Paper 12 / arXiv:2607.18768 Open arXiv

Relevance 5 Novelty 6

Why selected: Strong match for criterion 3: it introduces a new simulator/benchmark for embodied-style cross-modal tracking (CM-UOT) and a method tailored to a previously underexplored modality-switching setting.

Unmanned Aerial Vehicle (UAV) object tracking has emerged as a popular research field with broad practical applications. Modern UAVs are increasingly equipped with both visible light and thermal infrared sensors. However, due to constraints in communication bandwidth, computational resources and power consumption, current systems often activate one modality and switch between modalities to maintain robust tracking in complex scenarios. Such modality switch inevitably leads to significant appearance change and sudden spatial shift, posing great challenges for existing tracking algorithms. To handle this problem, we propose a novel State-Aware Representation Learning Approach called SARLA, which perceives the inconsistent modality states of current frame with template and last frame in the target representations to adapt to the sudden changes in both appearance and position, for robust cross-modal object tracking. In particular, we propose the Modality State Aware Representation Module (MSARM) and Spatial State Aware Representation Module (SSARM). MSARM guides the model to learn appearance correlation, bridging the modality gap, while SSARM models cross-frame spatial correlation to mitigate sudden spatial shift impacts. In addition, we design a spatial shift prediction loss to further handle the effects of spatial variation caused by modality switch. To promote the development of this research field, we establish a large-scale video benchmark called CM-UOT, which consists of 1079 cross-modal sequences with an average video length greater than 621 frames and encompasses over 671K frames in total. Extensive experiments on CM-UOT dataset demonstrate the superior performance of the proposed SARLA against 20 excellent tracking methods. The source code, datasets, and evaluation protocols associated with this work are publicly available at: https://github.com/hongsmile365/sarla-.

14 Mitigating Modality and Language-Style Gaps for Zero-Shot Video Moment Retrieval Jihyun Lee, Cheol-Ho Cho, Woojin Jun, Woojin Jeong, Jae-Pil Heo

Vision-Language Reasoning Multimodal LLM Benchmark & Evaluation Generative Modeling

ArXiv

Paper 14 / arXiv:2607.19027 Open arXiv

Relevance 6 Novelty 5

Why selected: Matches criterion 2 and criterion 1 moderately: zero-shot video moment retrieval with an MLLM reasoning stage and a method to mitigate modality/language-style gaps.

Zero-shot video moment retrieval aims to overcome the limitations of traditional approaches that require large-scale datasets annotated with text and its relevant temporal spans. Despite advances in pre-trained vision-language models and multimodal large language models, existing ZMR methods still heavily depend on query-to-video content similarity, making them vulnerable to modality and language-style gaps. These gaps lead to unreliable span proposals and unstable moment retrieval results. To address this issue, we propose Self-Similarity-based Moment Proposal and Scoring that instead exploits intrinsic relationships within videos, enabling robust span generation and scoring. By deriving self-similarity only from the video content, we circumvent the noisy and mismatched patterns of query-frame or query-caption similarities, thereby mitigating both modality and language-style gaps. Furthermore, we introduce a query-aware MLLM-based reasoning stage to further sharpen alignment between text and video. Extensive experiments demonstrate that Self-SiMS achieves state-of-the-art performance across ZMR benchmarks.

16 OPD-IAD: From Language Judgment to Industrial Anomaly Detection via On-Policy Self-Distillation Shuimu Chen, Jing Jin, Nan Su, Hongbo Xu, Zebang Cheng, Wenming Yang, Fei Ma, Guijin Wang

Vision-Language Reasoning Safety & Backdoors Retrieval & Grounding

ArXiv

Paper 16 / arXiv:2607.18850 Open arXiv

Relevance 5 Novelty 5

Why selected: Matches criterion 2 closely: an LVLM-based industrial anomaly detection method that uses language-guided visual anchoring for pixel-level localization.

Large vision-language models (LVLMs) have recently shown strong potential for industrial anomaly detection (IAD) by providing image-level anomaly judgments and interpretable defect reasoning. However, current LVLM-based IAD methods still struggle to produce precise pixel-level anomaly maps from generated language judgments. We aim to achieve precise pixel-level localization while using language as guidance rather than letting it dominate the visual response. Specifically, we propose \textbf{OPD-IAD}, an evidence-privileged dense on-policy self-distillation framework for LVLM-based IAD. OPD-IAD distills privileged defect evidence onto the model's own on-policy judgment trajectory, enabling the final generated judgment to be learned under dense supervision rather than treated only as a textual answer. The resulting judgment serves as a semantic condition for dense anomaly perception. To turn this condition into dense visual evidence, we introduce \textbf{Language-guided Visual Anchoring}, which uses a judgment reforward to re-encode the image and question under the final-judgment condition into semantic anchors and contrasts them with dense visual features through a contrastive heatmap head to generate anomaly maps. The language judgment therefore provides compact semantic guidance, while dense visual features remain the basis for pixel-level scoring, allowing language to guide anomaly localization without letting language quality directly dictate the pixel-level response. Extensive experiments show that OPD-IAD achieves the best overall performance among LVLM-based IAD methods, leading on most image-level, pixel-level, and QA metrics.

17 DobicVLM: Aligning Chest X-Ray Report Generation with Clinically-Grounded Programmatic Rewards via Group Relative Policy Optimization Thanni Adewuyi, Angelica Obayi, Andem Aniekan, Samuel Okoko, Angel Ezendu, Ephraim Usani, Ademide Animasaun, Philip Chibundu, Christian Maurice, Mary Donald Essien, Oluwaseun Odunsi, Oluwasegun Oguntuase, Abiodun Adereni

Vision-Language Reasoning Benchmark & Evaluation Medical AI Generative Modeling

ArXiv

Paper 17 / arXiv:2607.18988 Open arXiv

Relevance 5 Novelty 5

Why selected: Matches criterion 2 closely: a vision-language model for chest X-ray report generation, with RL-style alignment using clinically grounded programmatic rewards.

Medical imaging is a cornerstone of diagnostics, yet automated chest X-ray report generation struggles with structural adherence, anatomical completeness, and semantic faithfulness. We introduce DobicVLM, a vision-language model combining supervised fine-tuning on MedGemma-4B with Group Relative Policy Optimization (GRPO) and clinically-grounded programmatic rewards. Our approach uses interpretable, rule-based reward components; structural verification, anatomical checklist, semantic similarity, and length constraints to enforce clinical standards without neural reward models. Trained on 1,000 de-identified image-report pairs from a private clinical dataset (with ethics approval and compliance to local regulations), DobicVLM is evaluated via blinded expert review on 69 held-out cases. DobicVLM outperforms Gemini 2.5 Flash across the majority of criteria, achieving the highest impression accuracy (27.2%) and medical terminology (86.5%) compared to both Gemini 2.5 Flash and MedGemma 4B baselines, with minor trade-offs in completeness and referrals. This demonstrates GRPO's value for transparent alignment in resource-limited settings. Keywords: Vision-Language Models, Radiology Report Generation, Reinforcement Learning, Medical AI, GRPO

19 CoGoal3D: Collaborative 3D Object Detection with 3D-Aware Fusion and Refinement Zhihao Yang, Zhiyu Xiang, Peng Xu, Tianyu Pu, Kai Wang, Eryun Liu, Dongping Zhang, Yong Ding

Vision-Language Reasoning Benchmark & Evaluation Embodied Interaction Safety & Backdoors

ArXiv

Paper 19 / arXiv:2607.19036 Open arXiv

Relevance 4 Novelty 5

Why selected: Matches criterion 1 partly: novel 3D spatial misalignment handling for collaborative 3D object detection in V2X perception.

V2X collaborative object detection features overcoming the limitations of single-vehicle systems by aggregating environmental features from multiple collaborative agents. However, existing mainstream V2X perception methods mainly focus on 2D BEV object detection. When 3D detection task is concerned, inferior results are obtained because they ignore the 3D spatial misalignment caused by differing height and attitude among the collaborators. In this paper, we propose a novel collaborative 3D object detection framework called CoGoal3D, which extracts and refines the 3D feature gradually in a two-stage pipeline. In the first stage, a multiscale 3D-aware global fusion module is designed to mitigate the 3D spatial misalignment. The resulting proposals are then refined in the second stage with an auxiliary task of 3D point reconstruction. An effective multi-agent collaborative data augmentation strategy is further proposed to enrich the training data while minimizing information loss. Extensive experiments on public real-world datasets demonstrate that our CoGoal3D achieves new state-of-the-art performance, with 3D AP@0.7 improvements of 10.86%, 10.34%, and 10.18% on the DAIR-V2X, V2V4Real, and V2X-Real datasets, respectively. Code is available at https://github.com/Megalo-f/CoGoal3D.

21 CGMap: A Geospatially Aware Deep Learning Framework for Crop Gap Mapping Using UAV Karan Sharma, Rajiv Ranjan, Dinesh Kumar, Shashank Tamaskar

Vision-Language Reasoning Benchmark & Evaluation Safety & Backdoors

ArXiv

Paper 21 / arXiv:2607.18779 Open arXiv

Relevance 5 Novelty 3

Why selected: Matches criterion 4 only: a vision foundation model (YOLOv8 detector) is used for UAV-based crop gap mapping and geospatial field analysis.

In India, crop germination is primarily monitored by visual inspection and manual counting, which are prone to errors, despite their crucial role in determining eventual yield potential. This paper highlights a deep learning based pipeline which uses object detection methods and drone imagery to assess and provide a precise count of sugarcane germination in fields. The approch uses a pre-trained AI model to find germinated plant sampling and identify gaps, also known as ``bald spots'', which restricts field productivity. The techniques used here relies on the YOLOV8 architecture, which was trained on a carefully selected dataset of UAV photos taken in various agroclimatic zones of India. Here, we bring upon a novel orientation-normalization technique that uses minimum Spanning Trees (MST) to account for variations in planting geometry, allowing for dependable row and column extraction across a variety of field layouts. By converting detected seedlings into spatial point clouds, emergence gaps can be inferred from the anticipated spacing between plants. A geospatial germination map exported in Well-Known Text (WKT) format is the end result, and it can be easily incorporated into GIS platforms used by sugar mills and agronomists to direct transplant initiatives. Timely interventions based on the insights provided by the algorithm can significantly increase yield, resulting in higher profits. Hence, support proper allocation of resources, avoid wastage, and enhance long-term sustainability.

24 Benchmarking Deep Learning Approaches for AEC Engineering Drawing Layout Detection and Information Extraction Tianyang Huang, Alessio Lombardi, Ahmed Elnagar, Ahmed Zalouk, George Paul, Sepehr Najjarpour, Arvid Sigurdsson, Khalid Ismail, Mohamed Ragab, Edlira Vakaj

Vision-Language Reasoning Benchmark & Evaluation Safety & Backdoors

ArXiv

Paper 24 / arXiv:2607.18997 Open arXiv

Relevance 4 Novelty 3

Why selected: Matches criterion 4 and partly criterion 2: benchmarks layout detection/information extraction on AEC drawings and reports a VLM (Qwen3-VL) result on the task.

Information Extraction (IE) from Architecture, Engineering, and Construction (AEC) drawings remains hindered by manual inefficiency, while Layout Detection, a vital 'middleware' organizing graphical and textual hierarchies, is underexplored. General document layout models, optimized for text-centric content, lack validation on engineering drawings. This study constructs a custom AEC-specific layouts dataset and benchmarks five deep learning architectures. RF-DETR achieves state-of-the-art performance with an $mAP_{50}$ of 0.949, while the Vision-Language Model Qwen3-VL attains a leading F1-score of 0.911. Conversely, models pre-trained on general document datasets suffer from "domain interference", causing performance degradation. This establishes a robust technical foundation for automated IE in AEC.

Benchmark & Evaluation

3 Masked Visual Actions for Unified World Modeling Hadi Alzayer, Wenlong Huang, Haonan Chen, Christopher Luey, Lvmin Zhang, Maneesh Agrawala, Gordon Wetzstein, Li Fei-Fei, Yilun Du, Jiajun Wu, Jia-Bin Huang

Benchmark & Evaluation Embodied Interaction LLM Interpretability Retrieval & Grounding

ArXiv

Paper 3 / arXiv:2607.19343 Open arXiv

Relevance 6 Novelty 8

Why selected: Strong match for criterion 3: it proposes a new world-modeling method for embodied manipulation, with a pixel-space action interface and multi-embodiment control.

Video models absorb rich priors over how the visual world moves, interacts, and responds to contact, making them promising substrates for robotic world modeling. The central challenge is how to communicate action to such models in a form aligned with the visual space in which they learned these interaction priors, yet still grounded in physical manipulation. We introduce Masked Visual Actions, a pixel-space control interface that expresses action as a partially revealed trajectory of an arbitrary entity in a video. Revealing robot motion makes the model act as a forward dynamics model that predicts the scene's response to low-level robot actions, while revealing desired object motion makes the same model recover robot behavior consistent with that outcome. Finetuned with only 15 hours of masked examples from real videos and simulation, a single checkpoint achieves strong visual fidelity and controllability across diverse scenes and multiple embodiments. In downstream manipulation settings, the model produces imagined rollouts whose outcomes correlate with real-world execution for policy evaluation, improves decision making by ranking candidate futures in model-based planning, and supports inverse modeling by synthesizing robot motion from desired object motion.

6 Open-Vocabulary Gaze Object Prediction: Benchmark and Method Binglu Wang, Sensen Niu, Ying Chen, Guangyu Guo

Benchmark & Evaluation LLM Interpretability Retrieval & Grounding

ArXiv

Paper 6 / arXiv:2607.18827 Open arXiv

Relevance 7 Novelty 6

Why selected: Matches criterion 3 closely: it introduces a new benchmark for open-vocabulary gaze object prediction and a method tailored to in-the-wild gaze grounding.

Gaze Object Prediction (GOP) aims to localize and recognize the objects humans attend to, a task crucial for understanding human-centric interactions. However, existing methods are typically trained under a closed-vocabulary paradigm with a fixed label space and evaluated on scene-specific datasets, limiting their applicability to real-world scenarios where gaze targets often follow a long-tail distribution or belong to unseen categories. To address this gap, we introduce Diverse Scenes for Gaze object prediction (DiSG), a benchmark containing 86 in-the-wild categories that facilitates the evaluation of Open-Vocabulary GOP (OVGOP). Building on DiSG, we propose a framework that leverages text-driven object discovery to localize potential gaze candidates, with a gaze-guided selection module to pinpoint the intended target from the candidate objects. Furthermore, to better capture semantic knowledge across diverse in-the-wild categories, we introduce Gradient-Informed Selection Tuning (GIST) to selectively update parameters most relevant to a given class vocabulary. Extensive experiments demonstrate that our proposed model performs effectively in open-vocabulary settings and also outperforms existing methods in the conventional closed-vocabulary setting. The benchmark and code is available at https://github.com/sensniu/ovgop.

8 OmniReasoner: Thinking with Long Audio-Video via Native Tool Use Yu Chen, Caorui Li, Ziyu Xiong, Yidong Wang, Mingqi Gao, Shuman Liu, Biao Liu, Chunfeng Yang, Anxiang Zeng, Haibo Zhang, Chaofan Chen

Benchmark & Evaluation Time-Series LLM Retrieval & Grounding

ArXiv

Paper 8 / arXiv:2607.19339 Open arXiv

Relevance 6 Novelty 6

Why selected: Matches criterion 2 closely: an omnimodal VLLM/MLLM with native tool use for long audio-video reasoning and temporal grounding.

Long audio-video reasoning is difficult for omnimodal LLMs because the decisive evidence is often sparse, cross-modal, and too expensive to preserve with uniformly high-fidelity inputs. We introduce OmniReasoner, a tool-use post-training framework for Thinking with Long Audio-Video: omni-modal LLMs learn, via supervised fine-tuning and reinforcement learning, to decide whether and where to call a zoom-in tool before answering. OmniReasoner first builds a low-cost global preview of the full stream and then, when needed, calls the zoom-in tool with a requested temporal interval for higher-fidelity visual and audio inspection before answering. Because the model observes different sampling granularities before and after this call -- a sparse global preview and a denser local clip -- we introduce TimeAnchor, which keeps the tool's temporal argument valid and round-trip-consistent across these granularities, rather than tied to frame indices from a particular sampling rate. To make this tool-use behavior trainable without expensive manual interval annotation, we build a Temporal Augmented Data Engine that synthesizes tool-use post-training trajectories by video editing and composition. Experiments across omnimodal and video benchmarks show that OmniReasoner improves both answer accuracy and temporal grounding while concentrating high-fidelity computation on informative regions. Code is available at https://github.com/RockyChen0205/OmniReasoner.

10 Wave2Body: Rethinking mmWave Human Pose Estimation as Radar-to-Body Token Translation Bo Liang, Chen Gong, Wei Gao, Chenren Xu

Benchmark & Evaluation LLM Interpretability

ArXiv

Paper 10 / arXiv:2607.18875 Open arXiv

Relevance 6 Novelty 6

Why selected: Matches criterion 1 very closely: new methodological improvement for spatial understanding in embodied/human sensing, reformulated as radar-to-body token translation for 3D pose estimation.

Millimeter-wave (mmWave) radar enables privacy-friendly human sensing, but its sparse point clouds are physical measurements of view-dependent electromagnetic reflections and only indirectly characterize body articulation. Recovering a complete 3D pose from such partial, geometry-dependent observations is therefore under-constrained. Existing methods directly regress joint coordinates from paired radar-pose data, relying on the same limited paired supervision to learn radar perception, human-body structure, and their alignment. This coupling can encourage dataset-specific shortcuts under ambiguous radar observations. We propose Wave2Body, a radar-to-body token translation framework that decouples these learning targets using a self-supervised mmWave tokenizer, a pretrained compositional body tokenizer that defines the output space, and a lightweight translator between them. Experiments on M4Human and mmBody show that Wave2Body achieves stronger cross-domain generalization than previous methods while incurring much lower computational costs for training and inference. All the code and experiment results are publicly available at https://github.com/Galaxywalk/Wave2Body.

13 AniGS: Bridging Rendering and Diffusion Prior for 3D Scene Animation Yen-Chi Cheng, Chen Gao, Chuhan Chen, Tuotuo Li, Rajvi Shah, Ayush Saraf, Changil Kim, Liangyan Gui, Alexander Schwing, Johannes Kopf, Hung-Yu Tseng

Benchmark & Evaluation Generative Modeling

ArXiv

Paper 13 / arXiv:2607.18539 Open arXiv

Relevance 5 Novelty 6

Why selected: Matches criterion 4 closely: uses pretrained video diffusion models to animate 3D Gaussian Splatting scenes, a vision foundation model application.

Novel view rendering of large and complex reconstructed scenes is becoming increasingly photorealistic. However, most reconstructions remain static and lack the ambient motion that makes environments immersive. We present AniGS, a method for scene-level animation of 3D Gaussian Splatting (3DGS) reconstructions that adds subtle, distributed dynamics, e.g., vegetation motion, while preserving rigid structures. Unlike existing 3D animation techniques which are limited to object-centric subjects or small regions, AniGS is designed for large, cluttered, navigable scenes. AniGS represents the scene with a canonical 3DGS and models motion using a time-conditioned deformation field. To animate the entire scene, we leverage a pretrained video diffusion model and introduce an iterative dataset--model update strategy that progressively expands viewpoint coverage and repeatedly updates camera-fixed training videos using a render-and-refine scheme. To prevent artifacts from unintended motion in static areas, we further introduce a composed video-to-video refinement scheme that restricts motion to desired regions. Experiments on five real-world, large-scale outdoor scenes demonstrate that AniGS produces natural ambient dynamics and high-quality novel view videos, enabling more immersive viewing experiences of reconstructed environments.

20 Text-conditioned Segmentation for Tomato Phenotyping via Procedural Synthetic Data Samy Mounir, Mikolaj Cieslak, Najmeddine Dhieb, Hakim Ghazzai, Jonathan Klein, Katja Froehlich, Soeren Pirk, Wojciech Palubicki, Gianluca Setti, Ahmed M. Eltawil, Dominik L. Michels

Benchmark & Evaluation Generative Modeling

ArXiv

Paper 20 / arXiv:2607.18576 Open arXiv

Relevance 5 Novelty 4

Why selected: Matches criterion 4 closely: applies a foundation model (SAM 3) to text-conditioned segmentation with sim-to-real synthetic data for greenhouse phenotyping.

Vision-based automation is an excellent candidate for reducing manual labor in greenhouse crop production and phenotyping. However, progress is constrained by the lack of annotated training data. Recent advances in vision-based foundational models have shown promising results in zero-shot generalization to novel domains, but their performance drops in complex agricultural environments. In this work, we present a sim-to-real framework for tomato plant segmentation that combines synthetic data generation with fine-tuning of a foundation model. We model a commercial cherry tomato greenhouse and use it to generate a large-scale synthetic dataset under diverse viewpoints, lighting conditions, and plant morphology. Subsequently, we fine-tune the Segment Anything Model 3 (SAM 3) on the synthetic dataset, specializing its text-conditioned segmentation behavior for greenhouse crop organs while retaining the general visual prior that makes zero-shot transfer possible. By evaluating our framework on multiple real-world greenhouse datasets, we demonstrate that combining synthetic data with SAM 3 fine-tuning significantly improves segmentation performance and model confidence. To support community benchmarking, we publicly release the procedural model, the generated synthetic dataset, and our fine-tuned SAM 3 weights.

Generative Modeling

15 Moving Alphabet: A Controlled Study of Training Data for Text-to-Video Generation Amber Yijia Zheng, Lu Liu, Raymond A. Yeh, Xi Yin

Generative Modeling

ArXiv

Paper 15 / arXiv:2607.18789 Open arXiv

Relevance 4 Novelty 6

Why selected: Good match for criterion 4: it studies training data for text-to-video generation and provides controlled insights that are useful for video foundation models.

Text-to-video generation has advanced significantly over the past five years through scaling of model size, data, and compute. Unlike model architecture, training data is often underexplored. Real-world data curation is complex and non-trivial, involving clip selection from raw videos and captioning to create video-text pairs for learning text-to-video mappings. We study how data distribution and caption quality impact text-to-video models. To enable controlled experiments, we introduce Moving Alphabet, a procedural testbed that renders letters with varying fonts, colors, sizes, and positions, moving in different directions and speeds against a black background. This design allows precise control over data distribution and caption quality by corrupting ground-truth metadata. Our experiments yield three findings: a) a diverse and balanced distribution of video content and duration is critical for generalization; b) caption quality significantly affects both model performance and training efficiency, suggesting that text-to-video models are bounded by video understanding capabilities; and c) classifier-free guidance and fine-tuning on high-quality data provide partial recovery from models trained on corrupted captions, but cannot fully compensate for poor pre-training data. We believe these insights can inform the development of large-scale text-to-video models, and we advocate for greater attention to the science of pre-training data.

18 DuSPiT: Dual-Branch Sub-Patch Pixel Diffusion Transformer Yunpeng Bai, Yossi Gandelsman, Micha\"el Gharbi

Generative Modeling LLM Interpretability

ArXiv

Paper 18 / arXiv:2607.18510 Open arXiv

Relevance 4 Novelty 5

Why selected: Moderate match for criterion 4: it is a vision foundation model architecture for image generation, with a focus on pixel-space diffusion quality and efficiency.

Diffusion Transformers achieve strong image generation performance, but most operate in compressed latent spaces. Pixel-space diffusion avoids this information loss, yet existing approaches map each raw image patch to a single token, forcing one representation to handle both global communication and fine-grained details. We address this issue by proposing a new architecture, \textbf{DuSPiT}, a \textbf{Du}al-branch \textbf{S}ub\textbf{P}atch \textbf{Pi}xel \textbf{T}ransformer. This model separates global structural reasoning from local appearance modeling. DuSPiT uses a compact base branch for efficient global reasoning and a parallel, high-capacity pixel branch, organized into subpatch groups, to preserve detailed appearance, with the two branches interacting through cross-attention. Our results show that DuSPiT generates images with richer details and stronger fine-grained structures, while also achieving a better quality--efficiency trade-off than prior pixel-space diffusion transformers.

General ML

22 Point Ladder Tuning: Parameter-Efficient Hierarchical Adaptation for 3D Point Cloud Understanding Junlin Chang, Longhao Zou, Rui Li

General ML

ArXiv

Paper 22 / arXiv:2607.19171 Open arXiv

Relevance 3 Novelty 5

Why selected: Matches criterion 4 loosely: parameter-efficient adaptation for 3D point cloud backbones, relevant to foundation-model-style representation learning but not a vision foundation model application in the strict sense.

Fine-tuning pre-trained point-cloud backbones typically updates all parameters, resulting in substantial computation and memory overhead. More importantly, modern point backbones rely on aggressive tokenization and downsampling, which yields compact global tokens but irreversibly discards fine-grained local geometry, an inherent bottleneck for parameter-efficient adaptation. Consequently, existing PEFT methods that operate only on these coarsened tokens can modulate global semantics but struggle to recover the missing multi-scale locality. We present Point Ladder Tuning (PLT), a locality-aware PEFT framework that performs hierarchical, instance-conditioned adaptation while keeping the backbone frozen. PLT forms a lightweight closed loop: (i) a Hierarchical Ladder Network (HLN) constructs a multi-resolution local feature pyramid directly from raw points; (ii) a Local-Global Fusion (LGF) aligns and fuses local pyramids with intermediate backbone semantics; and (iii) a Dynamic Prompt Generator produces instance-aware multi-scale prompts to modulate the frozen backbone effectively. For dense prediction, we further introduce a lightweight segmentation head that progressively upsamples fused features and leverages backbone priors to refine fine structures. Extensive experiments on classification and dense prediction show that PLT consistently surpasses prior PEFT baselines with minimal tunable parameters. PLT achieves state-of-the-art performance using only 2.71% trainable parameters for classification and 7.69% for dense prediction, and scales favorably to larger backbones, requiring merely 0.36% parameters on PointGPT-L. The code is released at https://github.com/JunLinChang/ECCV2026-PLT.

Multimodal LLM

23 Mi-Memory: A Lifecycle Memory Framework for Personal AI Xule Liu, Hanlin Teng, Chao Li, Yanan Ni, Shuo Lu, Audrey Wang, Yijun Liu, Yunfei Wang, Xiaofeng Li, Xian Yi, Yuanfa Li, Kang Zhao, Jian Liang, Yuxuan Chen, Jinyuan Chen, Heng Qu, Kun Shao, Jian Luan

Multimodal LLM Benchmark & Evaluation Embodied Interaction LLM Interpretability

ArXiv

Paper 23 / arXiv:2607.18975 Open arXiv

Relevance 4 Novelty 4

Why selected: Matches criterion 3 loosely: an embodied/personal AI memory framework with multimodal device evidence and deployment-aware memory, though it is more of a system report than a new embodied benchmark or core method.

Personal AI is moving beyond chat-only interaction toward continuous services that span phones, cars, homes, wearables, cameras, and tools. In this setting, memory cannot remain a cache of prior conversations. It should serve as a continuity and governance substrate: preserving durable user state, grounding answers in multimodal and device evidence, supporting correction and forgetting, bounding policy evolution, and remaining deployable under latency, cost, privacy, and edge-cloud constraints. This technical report presents Mi-Memory, a lifecycle memory framework for Personal AI organized around four roles: Structure, Expansion, Evolution, and Deployment. A shared audit contract links these roles through four recurring artifact families: typed evidence payloads preserve source identity and provenance, diagnostic traces localize evidence loss across the serving pipeline, strategy artifacts make memory-policy changes explicit, and gate/rollback records bound accepted evolution. MiMemory instantiates the roles through MemStack, MemSense/MemFuse, D$^{2}$ACCI/E$^{2}$MEND, and LiteMem. In controlled-reference Structure evaluations, MemStack reaches 93.59%, 57.24%, and 87.47% on LoCoMo, PersonaMem-V2, and LongMemEval, respectively; other tracks report module-level, preliminary/internal, transfer-feasibility, or design-only evidence with explicit boundaries. MiMemory is a step toward auditable, evidence-gated, and deployment-aware memory systems for Personal AI. Project homepage: https://darwin-agent.github.io/Mi-Memory/ .

Archived papers 24

Top score 17

Average score 11.1