Total relevant papers: 44
Paper selection prompt and criteria at the bottom
Table of contents with paper titles:
VAR: Visual Attention Reasoning via Structured Search and Backtracking Authors: Wei Cai, Jian Zhao, Yuchen Yuan, Tianle Zhang, Ming Zhu, Haichuan Tang, Chi Zhang, Xuelong Li
DSI-Bench: A Benchmark for Dynamic Spatial Intelligence Authors: Ziang Zhang, Zehan Wang, Guanghao Zhang, Weilong Dai, Yan Xia, Ziang Yan, Minjie Hong, Zhou Zhao
SAM 2++: Tracking Anything at Any Granularity Authors: Jiaming Zhang, Cheng Liang, Yichun Yang, Chenkai Zeng, Yutao Cui, Xinwen Zhang, Xin Zhou, Kai Ma, Gangshan Wu, Limin Wang
World-in-World: World Models in a Closed-Loop World Authors: Jiahan Zhang, Muqing Jiang, Nanru Dai, Taiming Lu, Arda Uzunoglu, Shunchi Zhang, Yana Wei, Jiahao Wang, Vishal M. Patel, Paul Pu Liang, Daniel Khashabi, Cheng Peng, Rama Chellappa, Tianmin Shu, Alan Yuille, Yilun Du, Jieneng Chen
Earth AI: Unlocking Geospatial Insights with Foundation Models and Cross-Modal Reasoning Authors: Aaron Bell, Amit Aides, Amr Helmy, Arbaaz Muslim, Aviad Barzilai, Aviv Slobodkin, Bolous Jaber, David Schottlander, George Leifman, Joydeep Paul, Mimi Sun, Nadav Sherman, Natalie Williams, Per Bjornsson, Roy Lee, Ruth Alcantara, Thomas Turnbull, Tomer Shekel, Vered Silverman, Yotam Gigi, Adam Boulanger, Alex Ottenwess, Ali Ahmadalipour, Anna Carter, Charles Elliott, David Andre, Elad Aharoni, Gia Jung, Hassler Thurston, Jacob Bien, Jamie McPike, Juliet Rothenberg, Kartik Hegde, Kel Markert, Kim Philipp Jablonski, Luc Houriez, Monica Bharel, Phing VanLee, Reuven Sayag, Sebastian Pilarski, Shelley Cazares, Shlomi Pasternak, Siduo Jiang, Stone Jiang, Thomas Colthurst, Yang Chen, Yehonathan Refael, Yochai Blau, Yuval Carny, Yael Maguire, Avinatan Hassidim, James Manyika, Tim Thelin, Genady Beryozkin, Gautam Prasad, Luke Barrington, Yossi Matias, Niv Efron, Shravya Shetty
Online In-Context Distillation for Low-Resource Vision Language Models Authors: Zhiqi Kang, Rahaf Aljundi, Vaggelis Dorovatas, Karteek Alahari
GBlobs: Local LiDAR Geometry for Improved Sensor Placement Generalization Authors: Du\v{s}an Mali'c, Christian Fruhwirth-Reisinger, Alexander Prutsch, Wei Lin, Samuel Schulter, Horst Possegger
PLANA3R: Zero-shot Metric Planar 3D Reconstruction via Feed-Forward Planar Splatting Authors: Changkun Liu, Bin Tan, Zeran Ke, Shangzhan Zhang, Jiachen Liu, Ming Qian, Nan Xue, Yujun Shen, Tristan Braud
ProCLIP: Progressive Vision-Language Alignment via LLM-based Embedder Authors: Xiaoxing Hu, Kaicheng Yang, Ziyong Feng, Qi Ming, Zonghao Guo, Xiang An, Ziyong Feng, Junchi Yan, Xue Yang
Mono4DGS-HDR: High Dynamic Range 4D Gaussian Splatting from Alternating-exposure Monocular Videos Authors: Jinfeng Liu, Lingtong Kong, Mi Zhou, Jinwen Chen, Dan Xu
Vision Foundation Models Can Be Good Tokenizers for Latent Diffusion Models Authors: Tianci Bi, Xiaoyi Zhang, Yan Lu, Nanning Zheng
OmniNWM: Omniscient Driving Navigation World Models Authors: Bohan Li, Zhuang Ma, Dalong Du, Baorui Peng, Zhujin Liang, Zhenqiang Liu, Chao Ma, Yueming Jin, Hao Zhao, Wenjun Zeng, Xin Jin
3D Weakly Supervised Semantic Segmentation via Class-Aware and Geometry-Guided Pseudo-Label Refinement Authors: Xiaoxu Xu, Xuexun Liu, Jinlong Li, Yitian Yuan, Qiudan Zhang, Lin Ma, Nicu Sebe, Xu Wang
UniGenBench++: A Unified Semantic Evaluation Benchmark for Text-to-Image Generation Authors: Yibin Wang, Zhimin Li, Yuhang Zang, Jiazi Bu, Yujie Zhou, Yi Xin, Junjun He, Chunyu Wang, Qinglin Lu, Cheng Jin, Jiaqi Wang
VLSU: Mapping the Limits of Joint Multimodal Understanding for AI Safety Authors: Shruti Palaskar, Leon Gatys, Mona Abdelrahman, Mar Jacobo, Larry Lindsey, Rutika Moharir, Gunnar Lund, Yang Xu, Navid Shiee, Jeffrey Bigham, Charles Maalouf, Joseph Yitan Cheng
AgentChangeBench: A Multi-Dimensional Evaluation Framework for Goal-Shift Robustness in Conversational AI Authors: Manik Rana, Calissa Man, Anotida Expected Msiiwa, Jeffrey Paine, Kevin Zhu, Sunishchal Dev, Vasu Sharma, Ahan M R
CoIDO: Efficient Data Selection for Visual Instruction Tuning via Coupled Importance-Diversity Optimization Authors: Yichen Yan, Ming Zhong, Qi Zhu, Xiaoling Gu, Jinpeng Chen, Huan Li
Zero-Shot Vehicle Model Recognition via Text-Based Retrieval-Augmented Generation Authors: Wei-Chia Chang, Yan-Ann Chen
HouseTour: A Virtual Real Estate A(I)gent Authors: Ata \c{C}elen, Marc Pollefeys, Daniel Barath, Iro Armeni
UWBench: A Comprehensive Vision-Language Benchmark for Underwater Understanding Authors: Da Zhang, Chenggang Rong, Bingyu Li, Feiyu Wang, Zhiyuan Zhao, Junyu Gao, Xuelong Li
SOCIA-Nabla: Textual Gradient Meets Multi-Agent Orchestration for Automated Simulator Generation Authors: Yuncheng Hua, Sion Weatherhead, Mehdi Jafari, Hao Xue, Flora D. Salim
Chimera: Compositional Image Generation using Part-based Concepting Authors: Shivam Singh, Yiming Chen, Agneet Chatterjee, Amit Raj, James Hays, Yezhou Yang, Chitra Baral
OpenInsGaussian: Open-vocabulary Instance Gaussian Segmentation with Context-aware Cross-view Fusion Authors: Tianyu Huang, Runnan Chen, Dongting Hu, Fengming Huang, Mingming Gong, Tongliang Liu
CovMatch: Cross-Covariance Guided Multimodal Dataset Distillation with Trainable Text Encoder Authors: Yongmin Lee, Hye Won Chung
Beyond Single Images: Retrieval Self-Augmented Unsupervised Camouflaged Object Detection Authors: Ji Du, Xin Wang, Fangwei Hao, Mingyang Yu, Chunyuan Chen, Jiesheng Wu, Bin Wang, Jing Xu, Ping Li
ViSE: A Systematic Approach to Vision-Only Street-View Extrapolation Authors: Kaiyuan Tan, Yingying Shen, Haiyang Sun, Bing Wang, Guang Chen, Hangjun Ye
Planned Diffusion Authors: Daniel Israel, Tian Jin, Ellie Cheng, Guy Van den Broeck, Aditya Grover, Suvinay Subramanian, Michael Carbin
Shortcutting Pre-trained Flow Matching Diffusion Models is Almost Free Lunch Authors: Xu Cai, Yang Wu, Qianli Chen, Haoran Wu, Lichuan Xiang, Hongkai Wen
LAND: Lung and Nodule Diffusion for 3D Chest CT Synthesis with Anatomical Guidance Authors: Anna Oliveras, Roger Mar'i, Rafael Redondo, Oriol Guardi`a, Ana Tost, Bhalaji Nagarajan, Carolina Migliorelli, Vicent Ribas, Petia Radeva
Cross-Modal Scene Semantic Alignment for Image Complexity Assessment Authors: Yuqing Luo, Yixiao Li, Jiang Liu, Jun Fu, Hadi Amirpour, Guanghui Yue, Baoquan Zhao, Padraig Corcoran, Hantao Liu, Wei Zhou
MAT-Agent: Adaptive Multi-Agent Training Optimization Authors: Jusheng Zhang, Kaitong Cai, Yijia Fan, Ningyuan Liu, Keze Wang
ScaleNet: Scaling up Pretrained Neural Networks with Incremental Parameters Authors: Zhiwei Hao, Jianyuan Guo, Li Shen, Kai Han, Yehui Tang, Han Hu, Yunhe Wang
Illusions of reflection: open-ended task reveals systematic failures in Large Language Models' reflective reasoning Authors: Sion Weatherhead, Flora Salim, Aaron Belbasis
Enhancing Few-Shot Classification of Benchmark and Disaster Imagery with ATTBHFA-Net Authors: Gao Yu Lee, Tanmoy Dam, Md Meftahul Ferdaus, Daniel Puiu Poenar, Vu Duong
Ranking-based Preference Optimization for Diffusion Models from Implicit User Feedback Authors: Yi-Lun Wu, Bo-Kai Ruan, Chiang Tseng, Hong-Han Shuai
ssToken: Self-modulated and Semantic-aware Token Selection for LLM Fine-tuning Authors: Xiaohan Qin, Xiaoxing Wang, Ning Liao, Cancheng Zhang, Xiangdong Zhang, Mingquan Feng, Jingzhi Wang, Junchi Yan
Counterfactual Reasoning for Steerable Pluralistic Value Alignment of Large Language Models Authors: Hanze Guo, Jing Yao, Xiao Zhou, Xiaoyuan Yi, Xing Xie
Beyond Frequency: Scoring-Driven Debiasing for Object Detection via Blueprint-Prompted Image Synthesis Authors: Xinhao Cai, Liulei Li, Gensheng Pei, Tao Chen, Jinshan Pan, Yazhou Yao, Wenguan Wang
DP$^2$O-SR: Direct Perceptual Preference Optimization for Real-World Image Super-Resolution Authors: Rongyuan Wu, Lingchen Sun, Zhengqiang Zhang, Shihao Wang, Tianhe Wu, Qiaosi Yi, Shuai Li, Lei Zhang
Query Decomposition for RAG: Balancing Exploration-Exploitation Authors: Roxana Petcu, Kenton Murray, Daniel Khashabi, Evangelos Kanoulas, Maarten de Rijke, Dawn Lawrie, Kevin Duh
Learning from Generalization Patterns: An Evaluation-Driven Approach to Enhanced Data Augmentation for Fine-Tuning Small Language Models Authors: Huan Song, Deeksha Razdan, Yiyue Qian, Arijit Ghosh Chowdhury, Parth Patwa, Aman Chadha, Shinan Zhang, Sharlina Keshava, Hannah Marlowe
S2AP: Score-space Sharpness Minimization for Adversarial Pruning Authors: Giorgio Piras, Qi Zhao, Fabio Brau, Maura Pintor, Christian Wressnegger, Battista Biggio
From Volume Rendering to 3D Gaussian Splatting: Theory and Applications Authors: Vitor Pereira Matias, Daniel Perazzo, Vinicius Silva, Alberto Raposo, Luiz Velho, Afonso Paiva, Tiago Novello
Beyond More Context: Retrieval Diversity Boosts Multi-Turn Intent Understanding Authors: Zhiming Lin
ArXiv ID: 2510.18619 Authors: Wei Cai, Jian Zhao, Yuchen Yuan, Tianle Zhang, Ming Zhu, Haichuan Tang, Chi Zhang, Xuelong Li
Abstract: Multimodal Large Language Models (MLLMs), despite their advances, are hindered by their high hallucination tendency and heavy reliance on brittle, linear reasoning processes, leading to failures in complex tasks. To address these limitations, we introduce Visual Attention Reasoning (VAR), a novel framework that recasts grounded reasoning as a structured search over a reasoning trajectory space. VAR decomposes the reasoning process into two key stages: traceable evidence grounding and search-based chain-of-thought (CoT) generation, which incorporates a backtracking mechanism for self-correction. The search is guided by a multi-faceted reward function with semantic and geometric self-verification components, which penalize outputs that are not faithfully grounded in the visual input. We provide a theoretical analysis for our search strategy, validating its capability to find the correct solution with high probability. Experimental results show that our 7B model, VAR-7B, sets a new state-of-the-art on a comprehensive suite of hallucination and safety benchmarks, significantly outperforming existing open-source models and demonstrating competitive performance against leading proprietary systems.
Comment: Matches criterion 2 (Visual and Multimodal Large Language Models) as it introduces a novel framework for grounded reasoning in MLLMs with a focus on reducing hallucination. Relevance: 10 Novelty: 8
ArXiv ID: 2510.18873 Authors: Ziang Zhang, Zehan Wang, Guanghao Zhang, Weilong Dai, Yan Xia, Ziang Yan, Minjie Hong, Zhou Zhao
Abstract: Reasoning about dynamic spatial relationships is essential, as both observers and objects often move simultaneously. Although vision-language models (VLMs) and visual expertise models excel in 2D tasks and static scenarios, their ability to fully understand dynamic 3D scenarios remains limited. We introduce Dynamic Spatial Intelligence and propose DSI-Bench, a benchmark with nearly 1,000 dynamic videos and over 1,700 manually annotated questions covering nine decoupled motion patterns of observers and objects. Spatially and temporally symmetric designs reduce biases and enable systematic evaluation of models' reasoning about self-motion and object motion. Our evaluation of 14 VLMs and expert models reveals key limitations: models often conflate observer and object motion, exhibit semantic biases, and fail to accurately infer relative relationships in dynamic scenarios. Our DSI-Bench provides valuable findings and insights about the future development of general and expertise models with dynamic spatial intelligence.
Comment: Matches criterion 1 (Spatial Intelligence and Embodied Agents) as it introduces a benchmark for dynamic spatial reasoning in vision-language models. Relevance: 10 Novelty: 7
ArXiv ID: 2510.18822 Authors: Jiaming Zhang, Cheng Liang, Yichun Yang, Chenkai Zeng, Yutao Cui, Xinwen Zhang, Xin Zhou, Kai Ma, Gangshan Wu, Limin Wang
Abstract: Video tracking aims at finding the specific target in subsequent frames given its initial state. Due to the varying granularity of target states across different tasks, most existing trackers are tailored to a single task and heavily rely on custom-designed modules within the individual task, which limits their generalization and leads to redundancy in both model design and parameters. To unify video tracking tasks, we present SAM 2++, a unified model towards tracking at any granularity, including masks, boxes, and points. First, to extend target granularity, we design task-specific prompts to encode various task inputs into general prompt embeddings, and a unified decoder to unify diverse task results into a unified form pre-output. Next, to satisfy memory matching, the core operation of tracking, we introduce a task-adaptive memory mechanism that unifies memory across different granularities. Finally, we introduce a customized data engine to support tracking training at any granularity, producing a large and diverse video tracking dataset with rich annotations at three granularities, termed Tracking-Any-Granularity, which represents a comprehensive resource for training and benchmarking on unified tracking. Comprehensive experiments on multiple benchmarks confirm that SAM 2++ sets a new state of the art across diverse tracking tasks at different granularities, establishing a unified and robust tracking framework.
Comment: Matches criterion 6 (Video Understanding) as it introduces a unified model for video tracking at any granularity, including masks, boxes, and points. Relevance: 9 Novelty: 8
ArXiv ID: 2510.18135 Authors: Jiahan Zhang, Muqing Jiang, Nanru Dai, Taiming Lu, Arda Uzunoglu, Shunchi Zhang, Yana Wei, Jiahao Wang, Vishal M. Patel, Paul Pu Liang, Daniel Khashabi, Cheng Peng, Rama Chellappa, Tianmin Shu, Alan Yuille, Yilun Du, Jieneng Chen
Abstract: Generative world models (WMs) can now simulate worlds with striking visual realism, which naturally raises the question of whether they can endow embodied agents with predictive perception for decision making. Progress on this question has been limited by fragmented evaluation: most existing benchmarks adopt open-loop protocols that emphasize visual quality in isolation, leaving the core issue of embodied utility unresolved, i.e., do WMs actually help agents succeed at embodied tasks? To address this gap, we introduce World-in-World, the first open platform that benchmarks WMs in a closed-loop world that mirrors real agent-environment interactions. World-in-World provides a unified online planning strategy and a standardized action API, enabling heterogeneous WMs for decision making. We curate four closed-loop environments that rigorously evaluate diverse WMs, prioritize task success as the primary metric, and move beyond the common focus on visual quality; we also present the first data scaling law for world models in embodied settings. Our study uncovers three surprises: (1) visual quality alone does not guarantee task success, controllability matters more; (2) scaling post-training with action-observation data is more effective than upgrading the pretrained video generators; and (3) allocating more inference-time compute allows WMs to substantially improve closed-loop performance.
Comment: Matches criterion 3 (Embodied/Robotic AI: New Benchmarks or Methods) as it introduces a new benchmark, World-in-World, for evaluating world models in closed-loop embodied settings. Relevance: 9 Novelty: 8
ArXiv ID: 2510.18318 Authors: Aaron Bell, Amit Aides, Amr Helmy, Arbaaz Muslim, Aviad Barzilai, Aviv Slobodkin, Bolous Jaber, David Schottlander, George Leifman, Joydeep Paul, Mimi Sun, Nadav Sherman, Natalie Williams, Per Bjornsson, Roy Lee, Ruth Alcantara, Thomas Turnbull, Tomer Shekel, Vered Silverman, Yotam Gigi, Adam Boulanger, Alex Ottenwess, Ali Ahmadalipour, Anna Carter, Charles Elliott, David Andre, Elad Aharoni, Gia Jung, Hassler Thurston, Jacob Bien, Jamie McPike, Juliet Rothenberg, Kartik Hegde, Kel Markert, Kim Philipp Jablonski, Luc Houriez, Monica Bharel, Phing VanLee, Reuven Sayag, Sebastian Pilarski, Shelley Cazares, Shlomi Pasternak, Siduo Jiang, Stone Jiang, Thomas Colthurst, Yang Chen, Yehonathan Refael, Yochai Blau, Yuval Carny, Yael Maguire, Avinatan Hassidim, James Manyika, Tim Thelin, Genady Beryozkin, Gautam Prasad, Luke Barrington, Yossi Matias, Niv Efron, Shravya Shetty
Abstract: Geospatial data offers immense potential for understanding our planet. However, the sheer volume and diversity of this data along with its varied resolutions, timescales, and sparsity pose significant challenges for thorough analysis and interpretation. This paper introduces Earth AI, a family of geospatial AI models and agentic reasoning that enables significant advances in our ability to unlock novel and profound insights into our planet. This approach is built upon foundation models across three key domains--Planet-scale Imagery, Population, and Environment--and an intelligent Gemini-powered reasoning engine. We present rigorous benchmarks showcasing the power and novel capabilities of our foundation models and validate that when used together, they provide complementary value for geospatial inference and their synergies unlock superior predictive capabilities. To handle complex, multi-step queries, we developed a Gemini-powered agent that jointly reasons over our multiple foundation models along with large geospatial data sources and tools. On a new benchmark of real-world crisis scenarios, our agent demonstrates the ability to deliver critical and timely insights, effectively bridging the gap between raw geospatial data and actionable understanding.
Comment: Matches criterion 4 (Vision Foundation Models and Their Applications) as it discusses geospatial foundation models and their applications in multi-modal reasoning. Relevance: 8 Novelty: 7
ArXiv ID: 2510.18117 Authors: Zhiqi Kang, Rahaf Aljundi, Vaggelis Dorovatas, Karteek Alahari
Abstract: As the field continues its push for ever more resources, this work turns the spotlight on a critical question: how can vision-language models (VLMs) be adapted to thrive in low-resource, budget-constrained settings? While large VLMs offer strong performance, they are impractical to deploy in such settings. Small VLMs, on the other hand, are efficient but typically require costly fine-tuning to close the performance gap with larger models in the deployment domain. Inspired by the in-context learning framework, we propose an online In-Context Distillation (ICD) method, in which a small VLM collaborates with a stronger teacher model at inference time, distilling its knowledge via sparse demonstrations to efficiently bridge the gap between them. Our method is built on an in-depth analysis that identifies the scale and the choice of models for which vision-language ICL is currently feasible, and demonstrates the advantage of ICL over fine-tuning under constrained compute budgets. We enhance our method with a novel cross-modal demonstration selection strategy, teacher test-time scaling to reduce noise, and student uncertainty conditioning to dynamically populate a demonstration pool and minimize teacher queries. Our ICD method significantly boosts the performance of small models (up to 33%) using scarce teacher annotations (as low as 4%), and competes with the teacher's zero-shot performance.
Comment: Matches criterion 2 as it explores a novel method for improving low-resource vision-language models, which aligns with the interest in VLLMs and MLLMs. Relevance: 8 Novelty: 7
ArXiv ID: 2510.18539 Authors: Du\v{s}an Mali'c, Christian Fruhwirth-Reisinger, Alexander Prutsch, Wei Lin, Samuel Schulter, Horst Possegger
Abstract: This technical report outlines the top-ranking solution for RoboSense 2025: Track 3, achieving state-of-the-art performance on 3D object detection under various sensor placements. Our submission utilizes GBlobs, a local point cloud feature descriptor specifically designed to enhance model generalization across diverse LiDAR configurations. Current LiDAR-based 3D detectors often suffer from a \enquote{geometric shortcut} when trained on conventional global features (\ie, absolute Cartesian coordinates). This introduces a position bias that causes models to primarily rely on absolute object position rather than distinguishing shape and appearance characteristics. Although effective for in-domain data, this shortcut severely limits generalization when encountering different point distributions, such as those resulting from varying sensor placements. By using GBlobs as network input features, we effectively circumvent this geometric shortcut, compelling the network to learn robust, object-centric representations. This approach significantly enhances the model's ability to generalize, resulting in the exceptional performance demonstrated in this challenge.
Comment: Matches criterion 3 as it introduces a novel method (GBlobs) for improving generalization in 3D object detection, which is relevant to embodied/robotic AI. Relevance: 8 Novelty: 7
ArXiv ID: 2510.18714 Authors: Changkun Liu, Bin Tan, Zeran Ke, Shangzhan Zhang, Jiachen Liu, Ming Qian, Nan Xue, Yujun Shen, Tristan Braud
Abstract: This paper addresses metric 3D reconstruction of indoor scenes by exploiting their inherent geometric regularities with compact representations. Using planar 3D primitives - a well-suited representation for man-made environments - we introduce PLANA3R, a pose-free framework for metric Planar 3D Reconstruction from unposed two-view images. Our approach employs Vision Transformers to extract a set of sparse planar primitives, estimate relative camera poses, and supervise geometry learning via planar splatting, where gradients are propagated through high-resolution rendered depth and normal maps of primitives. Unlike prior feedforward methods that require 3D plane annotations during training, PLANA3R learns planar 3D structures without explicit plane supervision, enabling scalable training on large-scale stereo datasets using only depth and normal annotations. We validate PLANA3R on multiple indoor-scene datasets with metric supervision and demonstrate strong generalization to out-of-domain indoor environments across diverse tasks under metric evaluation protocols, including 3D surface reconstruction, depth estimation, and relative pose estimation. Furthermore, by formulating with planar 3D representation, our method emerges with the ability for accurate plane segmentation. The project page is available at https://lck666666.github.io/plana3r
Comment: This paper matches criterion 3 as it introduces a novel method (PLANA3R) for metric 3D reconstruction using planar splatting, which is relevant to embodied AI and 3D scene understanding. Relevance: 8 Novelty: 7
ArXiv ID: 2510.18795 Authors: Xiaoxing Hu, Kaicheng Yang, Ziyong Feng, Qi Ming, Zonghao Guo, Xiang An, Ziyong Feng, Junchi Yan, Xue Yang
Abstract: The original CLIP text encoder is limited by a maximum input length of 77 tokens, which hampers its ability to effectively process long texts and perform fine-grained semantic understanding. In addition, the CLIP text encoder lacks support for multilingual inputs. All these limitations significantly restrict its applicability across a broader range of tasks. Recent studies have attempted to replace the CLIP text encoder with an LLM-based embedder to enhance its ability in processing long texts, multilingual understanding, and fine-grained semantic comprehension. However, because the representation spaces of LLMs and the vision-language space of CLIP are pretrained independently without alignment priors, direct alignment using contrastive learning can disrupt the intrinsic vision-language alignment in the CLIP image encoder, leading to an underutilization of the knowledge acquired during pre-training. To address this challenge, we propose ProCLIP, a curriculum learning-based progressive vision-language alignment framework to effectively align the CLIP image encoder with an LLM-based embedder. Specifically, ProCLIP first distills knowledge from CLIP's text encoder into the LLM-based embedder to leverage CLIP's rich pretrained knowledge while establishing initial alignment between the LLM embedder and CLIP image encoder. Subsequently, ProCLIP further aligns the CLIP image encoder with the LLM-based embedder through image-text contrastive tuning, employing self-distillation regularization to avoid overfitting. To achieve a more effective alignment, instance semantic alignment loss and embedding structure alignment loss are employed during representation inheritance and contrastive tuning. The Code is available at https://github.com/VisionXLab/ProCLIP
Comment: Matches criterion 2 (Visual and Multimodal Large Language Models) as it proposes a progressive vision-language alignment framework using LLM-based embedders. Relevance: 8 Novelty: 7
ArXiv ID: 2510.18489 Authors: Jinfeng Liu, Lingtong Kong, Mi Zhou, Jinwen Chen, Dan Xu
Abstract: We introduce Mono4DGS-HDR, the first system for reconstructing renderable 4D high dynamic range (HDR) scenes from unposed monocular low dynamic range (LDR) videos captured with alternating exposures. To tackle such a challenging problem, we present a unified framework with two-stage optimization approach based on Gaussian Splatting. The first stage learns a video HDR Gaussian representation in orthographic camera coordinate space, eliminating the need for camera poses and enabling robust initial HDR video reconstruction. The second stage transforms video Gaussians into world space and jointly refines the world Gaussians with camera poses. Furthermore, we propose a temporal luminance regularization strategy to enhance the temporal consistency of the HDR appearance. Since our task has not been studied before, we construct a new evaluation benchmark using publicly available datasets for HDR video reconstruction. Extensive experiments demonstrate that Mono4DGS-HDR significantly outperforms alternative solutions adapted from state-of-the-art methods in both rendering quality and speed.
Comment: Matches criterion 6 (Video Understanding) as it introduces a system for reconstructing 4D HDR scenes from monocular videos, focusing on video-based tasks. Relevance: 8 Novelty: 7
ArXiv ID: 2510.18457 Authors: Tianci Bi, Xiaoyi Zhang, Yan Lu, Nanning Zheng
Abstract: The performance of Latent Diffusion Models (LDMs) is critically dependent on the quality of their visual tokenizer. While recent works have explored incorporating Vision Foundation Models (VFMs) via distillation, we identify a fundamental flaw in this approach: it inevitably weakens the robustness of alignment with the original VFM, causing the aligned latents to deviate semantically under distribution shifts. In this paper, we bypass distillation by proposing a more direct approach: Vision Foundation Model Variational Autoencoder (VFM-VAE). To resolve the inherent tension between the VFM's semantic focus and the need for pixel-level fidelity, we redesign the VFM-VAE decoder with Multi-Scale Latent Fusion and Progressive Resolution Reconstruction blocks, enabling high-quality reconstruction from spatially coarse VFM features. Furthermore, we provide a comprehensive analysis of representation dynamics during diffusion training, introducing the proposed SE-CKNNA metric as a more precise tool for this diagnosis. This analysis allows us to develop a joint tokenizer-diffusion alignment strategy that dramatically accelerates convergence. Our innovations in tokenizer design and training strategy lead to superior performance and efficiency: our system reaches a gFID (w/o CFG) of 2.20 in merely 80 epochs (a 10x speedup over prior tokenizers). With continued training to 640 epochs, it further attains a gFID (w/o CFG) of 1.62, establishing direct VFM integration as a superior paradigm for LDMs.
Comment: Matches criterion 4 (Vision Foundation Models and Their Applications) as it focuses on improving latent diffusion models by directly integrating vision foundation models. Relevance: 8 Novelty: 7
ArXiv ID: 2510.18313 Authors: Bohan Li, Zhuang Ma, Dalong Du, Baorui Peng, Zhujin Liang, Zhenqiang Liu, Chao Ma, Yueming Jin, Hao Zhao, Wenjun Zeng, Xin Jin
Abstract: Autonomous driving world models are expected to work effectively across three core dimensions: state, action, and reward. Existing models, however, are typically restricted to limited state modalities, short video sequences, imprecise action control, and a lack of reward awareness. In this paper, we introduce OmniNWM, an omniscient panoramic navigation world model that addresses all three dimensions within a unified framework. For state, OmniNWM jointly generates panoramic videos of RGB, semantics, metric depth, and 3D occupancy. A flexible forcing strategy enables high-quality long-horizon auto-regressive generation. For action, we introduce a normalized panoramic Plucker ray-map representation that encodes input trajectories into pixel-level signals, enabling highly precise and generalizable control over panoramic video generation. Regarding reward, we move beyond learning reward functions with external image-based models: instead, we leverage the generated 3D occupancy to directly define rule-based dense rewards for driving compliance and safety. Extensive experiments demonstrate that OmniNWM achieves state-of-the-art performance in video generation, control accuracy, and long-horizon stability, while providing a reliable closed-loop evaluation framework through occupancy-grounded rewards. Project page is available at https://github.com/Arlo0o/OmniNWM.
Comment: Matches criterion 3 (Embodied/Robotic AI: New Benchmarks or Methods) as it introduces a novel omniscient panoramic navigation world model for autonomous driving with state, action, and reward dimensions. Relevance: 8 Novelty: 7
ArXiv ID: 2510.17875 Authors: Xiaoxu Xu, Xuexun Liu, Jinlong Li, Yitian Yuan, Qiudan Zhang, Lin Ma, Nicu Sebe, Xu Wang
Abstract: 3D weakly supervised semantic segmentation (3D WSSS) aims to achieve semantic segmentation by leveraging sparse or low-cost annotated data, significantly reducing reliance on dense point-wise annotations. Previous works mainly employ class activation maps or pre-trained vision-language models to address this challenge. However, the low quality of pseudo-labels and the insufficient exploitation of 3D geometric priors jointly create significant technical bottlenecks in developing high-performance 3D WSSS models. In this paper, we propose a simple yet effective 3D weakly supervised semantic segmentation method that integrates 3D geometric priors into a class-aware guidance mechanism to generate high-fidelity pseudo labels. Concretely, our designed methodology first employs Class-Aware Label Refinement module to generate more balanced and accurate pseudo labels for semantic categrories. This initial refinement stage focuses on enhancing label quality through category-specific optimization. Subsequently, the Geometry-Aware Label Refinement component is developed, which strategically integrates implicit 3D geometric constraints to effectively filter out low-confidence pseudo labels that fail to comply with geometric plausibility. Moreover, to address the challenge of extensive unlabeled regions, we propose a Label Update strategy that integrates Self-Training to propagate labels into these areas. This iterative process continuously enhances pseudo-label quality while expanding label coverage, ultimately fostering the development of high-performance 3D WSSS models. Comprehensive experimental validation reveals that our proposed methodology achieves state-of-the-art performance on both ScanNet and S3DIS benchmarks while demonstrating remarkable generalization capability in unsupervised settings, maintaining competitive accuracy through its robust design.
Comment: Matches criterion 3 as it introduces a novel method for 3D weakly supervised semantic segmentation, which is relevant to embodied/robotic AI. Relevance: 7 Novelty: 7
ArXiv ID: 2510.18701 Authors: Yibin Wang, Zhimin Li, Yuhang Zang, Jiazi Bu, Yujie Zhou, Yi Xin, Junjun He, Chunyu Wang, Qinglin Lu, Cheng Jin, Jiaqi Wang
Abstract: Recent progress in text-to-image (T2I) generation underscores the importance of reliable benchmarks in evaluating how accurately generated images reflect the semantics of their textual prompt. However, (1) existing benchmarks lack the diversity of prompt scenarios and multilingual support, both essential for real-world applicability; (2) they offer only coarse evaluations across primary dimensions, covering a narrow range of sub-dimensions, and fall short in fine-grained sub-dimension assessment. To address these limitations, we introduce UniGenBench++, a unified semantic assessment benchmark for T2I generation. Specifically, it comprises 600 prompts organized hierarchically to ensure both coverage and efficiency: (1) spans across diverse real-world scenarios, i.e., 5 main prompt themes and 20 subthemes; (2) comprehensively probes T2I models' semantic consistency over 10 primary and 27 sub evaluation criteria, with each prompt assessing multiple testpoints. To rigorously assess model robustness to variations in language and prompt length, we provide both English and Chinese versions of each prompt in short and long forms. Leveraging the general world knowledge and fine-grained image understanding capabilities of a closed-source Multi-modal Large Language Model (MLLM), i.e., Gemini-2.5-Pro, an effective pipeline is developed for reliable benchmark construction and streamlined model assessment. Moreover, to further facilitate community use, we train a robust evaluation model that enables offline assessment of T2I model outputs. Through comprehensive benchmarking of both open- and closed-sourced T2I models, we systematically reveal their strengths and weaknesses across various aspects.
Comment: Matches criterion 7 (Vision-Focused Survey Papers) as it introduces a comprehensive benchmark for text-to-image generation, evaluating semantic consistency across diverse scenarios. Relevance: 8 Novelty: 6
ArXiv ID: 2510.18214 Authors: Shruti Palaskar, Leon Gatys, Mona Abdelrahman, Mar Jacobo, Larry Lindsey, Rutika Moharir, Gunnar Lund, Yang Xu, Navid Shiee, Jeffrey Bigham, Charles Maalouf, Joseph Yitan Cheng
Abstract: Safety evaluation of multimodal foundation models often treats vision and language inputs separately, missing risks from joint interpretation where benign content becomes harmful in combination. Existing approaches also fail to distinguish clearly unsafe content from borderline cases, leading to problematic over-blocking or under-refusal of genuinely harmful content. We present Vision Language Safety Understanding (VLSU), a comprehensive framework to systematically evaluate multimodal safety through fine-grained severity classification and combinatorial analysis across 17 distinct safety patterns. Using a multi-stage pipeline with real-world images and human annotation, we construct a large-scale benchmark of 8,187 samples spanning 15 harm categories. Our evaluation of eleven state-of-the-art models reveals systematic joint understanding failures: while models achieve 90%-plus accuracy on clear unimodal safety signals, performance degrades substantially to 20-55% when joint image-text reasoning is required to determine the safety label. Most critically, 34% of errors in joint image-text safety classification occur despite correct classification of the individual modalities, further demonstrating absent compositional reasoning capabilities. Additionally, we find that models struggle to balance refusing unsafe content while still responding to borderline cases that deserve engagement. For example, we find that instruction framing can reduce the over-blocking rate on borderline content from 62.4% to 10.4% in Gemini-1.5, but only at the cost of under-refusing on unsafe content with refusal rate dropping from 90.8% to 53.9%. Overall, our framework exposes weaknesses in joint image-text understanding and alignment gaps in current models, and provides a critical test bed to enable the next milestones in research on robust vision-language safety.
Comment: Matches criterion 2 as it evaluates multimodal foundation models and their safety, which is relevant to vision-language integration. Relevance: 7 Novelty: 6
ArXiv ID: 2510.18170 Authors: Manik Rana, Calissa Man, Anotida Expected Msiiwa, Jeffrey Paine, Kevin Zhu, Sunishchal Dev, Vasu Sharma, Ahan M R
Abstract: Goal changes are a defining feature of real world multi-turn interactions, yet current agent benchmarks primarily evaluate static objectives or one-shot tool use. We introduce AgentChangeBench, a benchmark explicitly designed to measure how tool augmented language model agents adapt to mid dialogue goal shifts across three enterprise domains. Our framework formalizes evaluation through four complementary metrics: Task Success Rate (TSR) for effectiveness, Tool Use Efficiency (TUE) for reliability, Tool Call Redundancy Rate (TCRR) for wasted effort, and Goal-Shift Recovery Time (GSRT) for adaptation latency. AgentChangeBench comprises 2,835 task sequences and five user personas, each designed to trigger realistic shift points in ongoing workflows. Using this setup, we evaluate several frontier models and uncover sharp contrasts obscured by traditional $\text{pass}@k$ scores: for example, GPT-4o reaches $92.2%$ recovery on airline booking shifts while Gemini collapses to $48.6%$, and retail tasks show near perfect parameter validity yet redundancy rates above $80%$, revealing major inefficiencies. These findings demonstrate that high raw accuracy does not imply robustness under dynamic goals, and that explicit measurement of recovery time and redundancy is essential. AgentChangeBench establishes a reproducible testbed for diagnosing and improving agent resilience in realistic enterprise settings.
Comment: This paper matches criterion 3 as it introduces a new benchmark (AgentChangeBench) for evaluating goal-shift robustness in conversational AI. Relevance: 7 Novelty: 6
ArXiv ID: 2510.17847 Authors: Yichen Yan, Ming Zhong, Qi Zhu, Xiaoling Gu, Jinpeng Chen, Huan Li
Abstract: Multimodal large language models (MLLMs) rely heavily on instruction tuning to align vision and language capabilities, yet the computational cost of training on large-scale datasets remains a major bottleneck. Existing data selection methods aim to mitigate this by selecting important and diverse subsets, but they often suffer from two critical drawbacks: high computational overhead from processing the entire dataset and suboptimal data selection due to separate treatment of importance and diversity. We introduce CoIDO, a novel dual-objective framework that jointly optimizes data importance and diversity to overcome these challenges. Unlike existing approaches that require costly evaluations across the whole dataset, CoIDO employs a lightweight plug-in scorer. This scorer is trained on just a small random sample of data to learn the distribution of the candidate set, drastically reducing computational demands. By leveraging a homoscedastic uncertainty-based formulation, CoIDO effectively balances importance and diversity during training, enabling efficient and scalable data selection. In our experiments, we trained the CoIDO scorer using only 20 percent of randomly sampled data. Once trained, CoIDO was applied to the entire dataset to select a 20 percent subset for instruction tuning. On the widely used LLaVA-1.5-7B model across ten downstream tasks, this selected subset achieved an impressive 98.2 percent of the performance of full-data fine-tuning, on average.
Comment: Matches criterion 2 (Visual and Multimodal Large Language Models) as it proposes a novel data selection framework for multimodal large language models to improve instruction tuning. Relevance: 7 Novelty: 6
ArXiv ID: 2510.18502 Authors: Wei-Chia Chang, Yan-Ann Chen
Abstract: Vehicle make and model recognition (VMMR) is an important task in intelligent transportation systems, but existing approaches struggle to adapt to newly released models. Contrastive Language-Image Pretraining (CLIP) provides strong visual-text alignment, yet its fixed pretrained weights limit performance without costly image-specific finetuning. We propose a pipeline that integrates vision language models (VLMs) with Retrieval-Augmented Generation (RAG) to support zero-shot recognition through text-based reasoning. A VLM converts vehicle images into descriptive attributes, which are compared against a database of textual features. Relevant entries are retrieved and combined with the description to form a prompt, and a language model (LM) infers the make and model. This design avoids large-scale retraining and enables rapid updates by adding textual descriptions of new vehicles. Experiments show that the proposed method improves recognition by nearly 20% over the CLIP baseline, demonstrating the potential of RAG-enhanced LM reasoning for scalable VMMR in smart-city applications.
Comment: Matches criterion 2 (Visual and Multimodal Large Language Models) as it integrates vision-language models with retrieval-augmented generation for zero-shot vehicle model recognition. Relevance: 7 Novelty: 6
ArXiv ID: 2510.18054 Authors: Ata \c{C}elen, Marc Pollefeys, Daniel Barath, Iro Armeni
Abstract: We introduce HouseTour, a method for spatially-aware 3D camera trajectory and natural language summary generation from a collection of images depicting an existing 3D space. Unlike existing vision-language models (VLMs), which struggle with geometric reasoning, our approach generates smooth video trajectories via a diffusion process constrained by known camera poses and integrates this information into the VLM for 3D-grounded descriptions. We synthesize the final video using 3D Gaussian splatting to render novel views along the trajectory. To support this task, we present the HouseTour dataset, which includes over 1,200 house-tour videos with camera poses, 3D reconstructions, and real estate descriptions. Experiments demonstrate that incorporating 3D camera trajectories into the text generation process improves performance over methods handling each task independently. We evaluate both individual and end-to-end performance, introducing a new joint metric. Our work enables automated, professional-quality video creation for real estate and touristic applications without requiring specialized expertise or equipment.
Comment: Matches criterion 1 as it focuses on spatial intelligence and 3D camera trajectory generation for embodied agents. Relevance: 5 Novelty: 7
ArXiv ID: 2510.18262 Authors: Da Zhang, Chenggang Rong, Bingyu Li, Feiyu Wang, Zhiyuan Zhao, Junyu Gao, Xuelong Li
Abstract: Large vision-language models (VLMs) have achieved remarkable success in natural scene understanding, yet their application to underwater environments remains largely unexplored. Underwater imagery presents unique challenges including severe light attenuation, color distortion, and suspended particle scattering, while requiring specialized knowledge of marine ecosystems and organism taxonomy. To bridge this gap, we introduce UWBench, a comprehensive benchmark specifically designed for underwater vision-language understanding. UWBench comprises 15,003 high-resolution underwater images captured across diverse aquatic environments, encompassing oceans, coral reefs, and deep-sea habitats. Each image is enriched with human-verified annotations including 15,281 object referring expressions that precisely describe marine organisms and underwater structures, and 124,983 question-answer pairs covering diverse reasoning capabilities from object recognition to ecological relationship understanding. The dataset captures rich variations in visibility, lighting conditions, and water turbidity, providing a realistic testbed for model evaluation. Based on UWBench, we establish three comprehensive benchmarks: detailed image captioning for generating ecologically informed scene descriptions, visual grounding for precise localization of marine organisms, and visual question answering for multimodal reasoning about underwater environments. Extensive experiments on state-of-the-art VLMs demonstrate that underwater understanding remains challenging, with substantial room for improvement. Our benchmark provides essential resources for advancing vision-language research in underwater contexts and supporting applications in marine science, ecological monitoring, and autonomous underwater exploration. Our code and benchmark will be available.
Comment: Matches criterion 6 as it introduces a new benchmark for underwater vision-language tasks, including captioning and question answering. Relevance: 5 Novelty: 7
ArXiv ID: 2510.18551 Authors: Yuncheng Hua, Sion Weatherhead, Mehdi Jafari, Hao Xue, Flora D. Salim
Abstract: In this paper, we present SOCIA-Nabla, an end-to-end, agentic framework that treats simulator construction asinstance optimization over code within a textual computation graph. Specialized LLM-driven agents are embedded as graph nodes, and a workflow manager executes a loss-driven loop: code synthesis -> execution -> evaluation -> code repair. The optimizer performs Textual-Gradient Descent (TGD), while human-in-the-loop interaction is reserved for task-spec confirmation, minimizing expert effort and keeping the code itself as the trainable object. Across three CPS tasks, i.e., User Modeling, Mask Adoption, and Personal Mobility, SOCIA-Nabla attains state-of-the-art overall accuracy. By unifying multi-agent orchestration with a loss-aligned optimization view, SOCIA-Nabla converts brittle prompt pipelines into reproducible, constraint-aware simulator code generation that scales across domains and simulation granularities. This work is under review, and we will release the code soon.
Comment: Matches criterion 3 as it introduces a novel simulator generation framework for embodied AI with multi-agent orchestration. Relevance: 5 Novelty: 7
ArXiv ID: 2510.18083 Authors: Shivam Singh, Yiming Chen, Agneet Chatterjee, Amit Raj, James Hays, Yezhou Yang, Chitra Baral
Abstract: Personalized image generative models are highly proficient at synthesizing images from text or a single image, yet they lack explicit control for composing objects from specific parts of multiple source images without user specified masks or annotations. To address this, we introduce Chimera, a personalized image generation model that generates novel objects by combining specified parts from different source images according to textual instructions. To train our model, we first construct a dataset from a taxonomy built on 464 unique (part, subject) pairs, which we term semantic atoms. From this, we generate 37k prompts and synthesize the corresponding images with a high-fidelity text-to-image model. We train a custom diffusion prior model with part-conditional guidance, which steers the image-conditioning features to enforce both semantic identity and spatial layout. We also introduce an objective metric PartEval to assess the fidelity and compositional accuracy of generation pipelines. Human evaluations and our proposed metric show that Chimera outperforms other baselines by 14% in part alignment and compositional accuracy and 21% in visual quality.
Comment: Matches criterion 5 as it focuses on compositional image generation using part-based concepts and textual instructions. Relevance: 5 Novelty: 6
ArXiv ID: 2510.18253 Authors: Tianyu Huang, Runnan Chen, Dongting Hu, Fengming Huang, Mingming Gong, Tongliang Liu
Abstract: Understanding 3D scenes is pivotal for autonomous driving, robotics, and augmented reality. Recent semantic Gaussian Splatting approaches leverage large-scale 2D vision models to project 2D semantic features onto 3D scenes. However, they suffer from two major limitations: (1) insufficient contextual cues for individual masks during preprocessing and (2) inconsistencies and missing details when fusing multi-view features from these 2D models. In this paper, we introduce \textbf{OpenInsGaussian}, an \textbf{Open}-vocabulary \textbf{Ins}tance \textbf{Gaussian} segmentation framework with Context-aware Cross-view Fusion. Our method consists of two modules: Context-Aware Feature Extraction, which augments each mask with rich semantic context, and Attention-Driven Feature Aggregation, which selectively fuses multi-view features to mitigate alignment errors and incompleteness. Through extensive experiments on benchmark datasets, OpenInsGaussian achieves state-of-the-art results in open-vocabulary 3D Gaussian segmentation, outperforming existing baselines by a large margin. These findings underscore the robustness and generality of our proposed approach, marking a significant step forward in 3D scene understanding and its practical deployment across diverse real-world scenarios.
Comment: Matches criterion 4 as it focuses on 3D scene understanding using vision foundation models and introduces improvements in multi-view feature fusion. Relevance: 5 Novelty: 6
ArXiv ID: 2510.18583 Authors: Yongmin Lee, Hye Won Chung
Abstract: Multimodal dataset distillation aims to synthesize a small set of image-text pairs that enables efficient training of large-scale vision-language models. While dataset distillation has shown promise in unimodal tasks, extending it to multimodal contrastive learning presents key challenges: learning cross-modal alignment and managing the high computational cost of large encoders. Prior approaches address scalability by freezing the text encoder and update only the image encoder and text projection layer. However, we find this severely limits semantic alignment and becomes a bottleneck for performance scaling. We propose CovMatch, a scalable dataset distillation framework that aligns the cross-covariance of real and synthetic features while regularizing feature distributions within each modality. Unlike prior approaches, CovMatch enables joint optimization of both encoders, leading to stronger cross-modal alignment and improved performance. Evaluated on Flickr30K and COCO, CovMatch outperforms state-of-the-art multimodal distillation methods and achieves up to 6.8% absolute gains in retrieval accuracy using only 500 synthetic pairs.
Comment: Matches criterion 5 as it focuses on combining image-text pairs and improving cross-modal alignment for vision-language models. Relevance: 5 Novelty: 6
ArXiv ID: 2510.18437 Authors: Ji Du, Xin Wang, Fangwei Hao, Mingyang Yu, Chunyuan Chen, Jiesheng Wu, Bin Wang, Jing Xu, Ping Li
Abstract: At the core of Camouflaged Object Detection (COD) lies segmenting objects from their highly similar surroundings. Previous efforts navigate this challenge primarily through image-level modeling or annotation-based optimization. Despite advancing considerably, this commonplace practice hardly taps valuable dataset-level contextual information or relies on laborious annotations. In this paper, we propose RISE, a RetrIeval SElf-augmented paradigm that exploits the entire training dataset to generate pseudo-labels for single images, which could be used to train COD models. RISE begins by constructing prototype libraries for environments and camouflaged objects using training images (without ground truth), followed by K-Nearest Neighbor (KNN) retrieval to generate pseudo-masks for each image based on these libraries. It is important to recognize that using only training images without annotations exerts a pronounced challenge in crafting high-quality prototype libraries. In this light, we introduce a Clustering-then-Retrieval (CR) strategy, where coarse masks are first generated through clustering, facilitating subsequent histogram-based image filtering and cross-category retrieval to produce high-confidence prototypes. In the KNN retrieval stage, to alleviate the effect of artifacts in feature maps, we propose Multi-View KNN Retrieval (MVKR), which integrates retrieval results from diverse views to produce more robust and precise pseudo-masks. Extensive experiments demonstrate that RISE outperforms state-of-the-art unsupervised and prompt-based methods. Code is available at https://github.com/xiaohainku/RISE.
Comment: Does not match any specific criteria but introduces an unsupervised method for camouflaged object detection, which may be of general interest in computer vision. Relevance: 3 Novelty: 6
ArXiv ID: 2510.18341 Authors: Kaiyuan Tan, Yingying Shen, Haiyang Sun, Bing Wang, Guang Chen, Hangjun Ye
Abstract: Realistic view extrapolation is critical for closed-loop simulation in autonomous driving, yet it remains a significant challenge for current Novel View Synthesis (NVS) methods, which often produce distorted and inconsistent images beyond the original trajectory. This report presents our winning solution which ctook first place in the RealADSim Workshop NVS track at ICCV 2025. To address the core challenges of street view extrapolation, we introduce a comprehensive four-stage pipeline. First, we employ a data-driven initialization strategy to generate a robust pseudo-LiDAR point cloud, avoiding local minima. Second, we inject strong geometric priors by modeling the road surface with a novel dimension-reduced SDF termed 2D-SDF. Third, we leverage a generative prior to create pseudo ground truth for extrapolated viewpoints, providing auxilary supervision. Finally, a data-driven adaptation network removes time-specific artifacts. On the RealADSim-NVS benchmark, our method achieves a final score of 0.441, ranking first among all participants.
Comment: Does not match any specific criteria but discusses a novel pipeline for street-view extrapolation, which may be of general interest in computer vision. Relevance: 3 Novelty: 6
ArXiv ID: 2510.18087 Authors: Daniel Israel, Tian Jin, Ellie Cheng, Guy Van den Broeck, Aditya Grover, Suvinay Subramanian, Michael Carbin
Abstract: A central challenge in large language model inference is the trade-off between generation speed and output quality. Autoregressive models produce high-quality text but generate tokens sequentially. Diffusion models can generate tokens in parallel but often need many iterations to match the same quality. We propose planned diffusion, a hybrid method that combines the strengths of both paradigms. Planned diffusion works in two stages: first, the model creates a short autoregressive plan that breaks the output into smaller, independent spans. Second, the model generates these spans simultaneously using diffusion. This approach expands the speed-quality Pareto frontier and provides a practical path to faster, high-quality text generation. On AlpacaEval, a suite of 805 instruction-following prompts, planned diffusion achieves Pareto-optimal trade-off between quality and latency, achieving 1.27x to 1.81x speedup over autoregressive generation with only 0.87% to 5.4% drop in win rate, respectively. Our sensitivity analysis shows that the planning mechanism of planned diffusion is minimal and reliable, and simple runtime knobs exist to provide flexible control of the quality-latency trade-off.
Comment: Does not match any specific criteria but discusses a novel hybrid method for text generation, which may be of general interest. Relevance: 3 Novelty: 6
ArXiv ID: 2510.17858 Authors: Xu Cai, Yang Wu, Qianli Chen, Haoran Wu, Lichuan Xiang, Hongkai Wen
Abstract: We present an ultra-efficient post-training method for shortcutting large-scale pre-trained flow matching diffusion models into efficient few-step samplers, enabled by novel velocity field self-distillation. While shortcutting in flow matching, originally introduced by shortcut models, offers flexible trajectory-skipping capabilities, it requires a specialized step-size embedding incompatible with existing models unless retraining from scratch$\unicode{x2013}$a process nearly as costly as pretraining itself. Our key contribution is thus imparting a more aggressive shortcut mechanism to standard flow matching models (e.g., Flux), leveraging a unique distillation principle that obviates the need for step-size embedding. Working on the velocity field rather than sample space and learning rapidly from self-guided distillation in an online manner, our approach trains efficiently, e.g., producing a 3-step Flux less than one A100 day. Beyond distillation, our method can be incorporated into the pretraining stage itself, yielding models that inherently learn efficient, few-step flows without compromising quality. This capability also enables, to our knowledge, the first few-shot distillation method (e.g., 10 text-image pairs) for dozen-billion-parameter diffusion models, delivering state-of-the-art performance at almost free cost.
Comment: Does not closely match any specific criterion but is generally relevant to diffusion models and efficient sampling techniques. Relevance: 3 Novelty: 6
ArXiv ID: 2510.18446 Authors: Anna Oliveras, Roger Mar'i, Rafael Redondo, Oriol Guardi`a, Ana Tost, Bhalaji Nagarajan, Carolina Migliorelli, Vicent Ribas, Petia Radeva
Abstract: This work introduces a new latent diffusion model to generate high-quality 3D chest CT scans conditioned on 3D anatomical masks. The method synthesizes volumetric images of size 256x256x256 at 1 mm isotropic resolution using a single mid-range GPU, significantly lowering the computational cost compared to existing approaches. The conditioning masks delineate lung and nodule regions, enabling precise control over the output anatomical features. Experimental results demonstrate that conditioning solely on nodule masks leads to anatomically incorrect outputs, highlighting the importance of incorporating global lung structure for accurate conditional synthesis. The proposed approach supports the generation of diverse CT volumes with and without lung nodules of varying attributes, providing a valuable tool for training AI models or healthcare professionals.
Comment: Does not match any specific criterion but is generally relevant to medical imaging and generative modeling. Relevance: 3 Novelty: 6
ArXiv ID: 2510.18377 Authors: Yuqing Luo, Yixiao Li, Jiang Liu, Jun Fu, Hadi Amirpour, Guanghui Yue, Baoquan Zhao, Padraig Corcoran, Hantao Liu, Wei Zhou
Abstract: Image complexity assessment (ICA) is a challenging task in perceptual evaluation due to the subjective nature of human perception and the inherent semantic diversity in real-world images. Existing ICA methods predominantly rely on hand-crafted or shallow convolutional neural network-based features of a single visual modality, which are insufficient to fully capture the perceived representations closely related to image complexity. Recently, cross-modal scene semantic information has been shown to play a crucial role in various computer vision tasks, particularly those involving perceptual understanding. However, the exploration of cross-modal scene semantic information in the context of ICA remains unaddressed. Therefore, in this paper, we propose a novel ICA method called Cross-Modal Scene Semantic Alignment (CM-SSA), which leverages scene semantic alignment from a cross-modal perspective to enhance ICA performance, enabling complexity predictions to be more consistent with subjective human perception. Specifically, the proposed CM-SSA consists of a complexity regression branch and a scene semantic alignment branch. The complexity regression branch estimates image complexity levels under the guidance of the scene semantic alignment branch, while the scene semantic alignment branch is used to align images with corresponding text prompts that convey rich scene semantic information by pair-wise learning. Extensive experiments on several ICA datasets demonstrate that the proposed CM-SSA significantly outperforms state-of-the-art approaches. Codes are available at https://github.com/XQ2K/First-Cross-Model-ICA.
Comment: Does not match any specific criterion but is generally relevant to cross-modal understanding in computer vision. Relevance: 3 Novelty: 6
ArXiv ID: 2510.17845 Authors: Jusheng Zhang, Kaitong Cai, Yijia Fan, Ningyuan Liu, Keze Wang
Abstract: Multi-label image classification demands adaptive training strategies to navigate complex, evolving visual-semantic landscapes, yet conventional methods rely on static configurations that falter in dynamic settings. We propose MAT-Agent, a novel multi-agent framework that reimagines training as a collaborative, real-time optimization process. By deploying autonomous agents to dynamically tune data augmentation, optimizers, learning rates, and loss functions, MAT-Agent leverages non-stationary multi-armed bandit algorithms to balance exploration and exploitation, guided by a composite reward harmonizing accuracy, rare-class performance, and training stability. Enhanced with dual-rate exponential moving average smoothing and mixed-precision training, it ensures robustness and efficiency. Extensive experiments across Pascal VOC, COCO, and VG-256 demonstrate MAT-Agent's superiority: it achieves an mAP of 97.4 (vs. 96.2 for PAT-T), OF1 of 92.3, and CF1 of 91.4 on Pascal VOC; an mAP of 92.8 (vs. 92.0 for HSQ-CvN), OF1 of 88.2, and CF1 of 87.1 on COCO; and an mAP of 60.9, OF1 of 70.8, and CF1 of 61.1 on VG-256. With accelerated convergence and robust cross-domain generalization, MAT-Agent offers a scalable, intelligent solution for optimizing complex visual models, paving the way for adaptive deep learning advancements.
Comment: Does not match any specific criterion but is generally relevant to adaptive training strategies in computer vision. Relevance: 3 Novelty: 6
ArXiv ID: 2510.18431 Authors: Zhiwei Hao, Jianyuan Guo, Li Shen, Kai Han, Yehui Tang, Han Hu, Yunhe Wang
Abstract: Recent advancements in vision transformers (ViTs) have demonstrated that larger models often achieve superior performance. However, training these models remains computationally intensive and costly. To address this challenge, we introduce ScaleNet, an efficient approach for scaling ViT models. Unlike conventional training from scratch, ScaleNet facilitates rapid model expansion with negligible increases in parameters, building on existing pretrained models. This offers a cost-effective solution for scaling up ViTs. Specifically, ScaleNet achieves model expansion by inserting additional layers into pretrained ViTs, utilizing layer-wise weight sharing to maintain parameters efficiency. Each added layer shares its parameter tensor with a corresponding layer from the pretrained model. To mitigate potential performance degradation due to shared weights, ScaleNet introduces a small set of adjustment parameters for each layer. These adjustment parameters are implemented through parallel adapter modules, ensuring that each instance of the shared parameter tensor remains distinct and optimized for its specific function. Experiments on the ImageNet-1K dataset demonstrate that ScaleNet enables efficient expansion of ViT models. With a 2$\times$ depth-scaled DeiT-Base model, ScaleNet achieves a 7.42% accuracy improvement over training from scratch while requiring only one-third of the training epochs, highlighting its efficiency in scaling ViTs. Beyond image classification, our method shows significant potential for application in downstream vision areas, as evidenced by the validation in object detection task.
Comment: Does not match any specific criterion but is generally relevant to computer vision and model scaling. Relevance: 3 Novelty: 6
ArXiv ID: 2510.18254 Authors: Sion Weatherhead, Flora Salim, Aaron Belbasis
Abstract: Humans do not just find mistakes after the fact -- we often catch them mid-stream because 'reflection' is tied to the goal and its constraints. Today's large language models produce reasoning tokens and 'reflective' text, but is it functionally equivalent with human reflective reasoning? Prior work on closed-ended tasks -- with clear, external 'correctness' signals -- can make 'reflection' look effective while masking limits in self-correction. We therefore test eight frontier models on a simple, real-world task that is open-ended yet rule-constrained, with auditable success criteria: to produce valid scientific test items, then revise after considering their own critique. First-pass performance is poor (often zero valid items out of 4 required; mean $\approx$ 1), and reflection yields only modest gains (also $\approx$ 1). Crucially, the second attempt frequently repeats the same violation of constraint, indicating 'corrective gains' arise largely from chance production of a valid item rather than error detection and principled, constraint-sensitive repair. Performance before and after reflection deteriorates as open-endedness increases, and models marketed for 'reasoning' show no advantage. Our results suggest that current LLM 'reflection' lacks functional evidence of the active, goal-driven monitoring that helps humans respect constraints even on a first pass. Until such mechanisms are instantiated in the model itself, reliable performance requires external structure that enforces constraints.
Comment: Does not match any specific criteria but provides insights into the limitations of reflective reasoning in LLMs, which may be of general interest. Relevance: 3 Novelty: 5
ArXiv ID: 2510.18326 Authors: Gao Yu Lee, Tanmoy Dam, Md Meftahul Ferdaus, Daniel Puiu Poenar, Vu Duong
Abstract: The increasing frequency of natural and human-induced disasters necessitates advanced visual recognition techniques capable of analyzing critical photographic data. With progress in artificial intelligence and resilient computational systems, rapid and accurate disaster classification has become crucial for efficient rescue operations. However, visual recognition in disaster contexts faces significant challenges due to limited and diverse data from the difficulties in collecting and curating comprehensive, high-quality disaster imagery. Few-Shot Learning (FSL) provides a promising approach to data scarcity, yet current FSL research mainly relies on generic benchmark datasets lacking remote-sensing disaster imagery, limiting its practical effectiveness. Moreover, disaster images exhibit high intra-class variation and inter-class similarity, hindering the performance of conventional metric-based FSL methods. To address these issues, this paper introduces the Attention-based Bhattacharyya-Hellinger Feature Aggregation Network (ATTBHFA-Net), which linearly combines the Bhattacharyya coefficient and Hellinger distances to compare and aggregate feature probability distributions for robust prototype formation. The Bhattacharyya coefficient serves as a contrastive margin that enhances inter-class separability, while the Hellinger distance regularizes same-class alignment. This framework parallels contrastive learning but operates over probability distributions rather than embedded feature points. Furthermore, a Bhattacharyya-Hellinger distance-based contrastive loss is proposed as a distributional counterpart to cosine similarity loss, used jointly with categorical cross-entropy to significantly improve FSL performance. Experiments on four FSL benchmarks and two disaster image datasets demonstrate the superior effectiveness and generalization of ATTBHFA-Net compared to existing approaches.
Comment: Does not closely match any specific criterion but is generally relevant to few-shot learning and disaster imagery classification. Relevance: 3 Novelty: 5
ArXiv ID: 2510.18353 Authors: Yi-Lun Wu, Bo-Kai Ruan, Chiang Tseng, Hong-Han Shuai
Abstract: Direct preference optimization (DPO) methods have shown strong potential in aligning text-to-image diffusion models with human preferences by training on paired comparisons. These methods improve training stability by avoiding the REINFORCE algorithm but still struggle with challenges such as accurately estimating image probabilities due to the non-linear nature of the sigmoid function and the limited diversity of offline datasets. In this paper, we introduce Diffusion Denoising Ranking Optimization (Diffusion-DRO), a new preference learning framework grounded in inverse reinforcement learning. Diffusion-DRO removes the dependency on a reward model by casting preference learning as a ranking problem, thereby simplifying the training objective into a denoising formulation and overcoming the non-linear estimation issues found in prior methods. Moreover, Diffusion-DRO uniquely integrates offline expert demonstrations with online policy-generated negative samples, enabling it to effectively capture human preferences while addressing the limitations of offline data. Comprehensive experiments show that Diffusion-DRO delivers improved generation quality across a range of challenging and unseen prompts, outperforming state-of-the-art baselines in both both quantitative metrics and user studies. Our source code and pre-trained models are available at https://github.com/basiclab/DiffusionDRO.
Comment: Does not closely match any specific criterion but is generally relevant to generative modeling and user preference optimization. Relevance: 3 Novelty: 5
ArXiv ID: 2510.18250 Authors: Xiaohan Qin, Xiaoxing Wang, Ning Liao, Cancheng Zhang, Xiangdong Zhang, Mingquan Feng, Jingzhi Wang, Junchi Yan
Abstract: Data quality plays a critical role in enhancing supervised fine-tuning (SFT) for large language models (LLMs), and token-level data selection has emerged as a promising direction for its fine-grained nature. Despite their strong empirical performance, existing token-level selection methods share two key limitations: (1) requiring training or accessing an additional reference model, and (2) relying solely on loss information for token selection, which cannot well preserve semantically important tokens that are not favored by loss-based metrics. To address these challenges, we propose ssToken, a Self-modulated and Semantic-aware Token Selection approach. ssToken leverages readily accessible history models to compute the per-token loss difference with the current model, which serves as a self-modulated signal that enables the model to adaptively select tokens along its optimization trajectory, rather than relying on excess loss from an offline-trained reference model as in prior works. We further introduce a semantic-aware, attention-based token importance estimation metric, orthogonal to loss-based selection and providing complementary semantic information for more effective filtering. Extensive experiments across different model families and scales demonstrate that both self-modulated selection and semantic-aware selection alone outperform full-data fine-tuning, while their integration--ssToken--achieves synergistic gains and further surpasses prior token-level selection methods, delivering performance improvements while maintaining training efficiency.
Comment: Does not closely match any specific criterion but is generally relevant to large language model fine-tuning. Relevance: 3 Novelty: 5
ArXiv ID: 2510.18526 Authors: Hanze Guo, Jing Yao, Xiao Zhou, Xiaoyuan Yi, Xing Xie
Abstract: As large language models (LLMs) become increasingly integrated into applications serving users across diverse cultures, communities and demographics, it is critical to align LLMs with pluralistic human values beyond average principles (e.g., HHH). In psychological and social value theories such as Schwartz's Value Theory, pluralistic values are represented by multiple value dimensions paired with various priorities. However, existing methods encounter two challenges when aligning with such fine-grained value objectives: 1) they often treat multiple values as independent and equally important, ignoring their interdependence and relative priorities (value complexity); 2) they struggle to precisely control nuanced value priorities, especially those underrepresented ones (value steerability). To handle these challenges, we propose COUPLE, a COUnterfactual reasoning framework for PLuralistic valuE alignment. It introduces a structural causal model (SCM) to feature complex interdependency and prioritization among features, as well as the causal relationship between high-level value dimensions and behaviors. Moreover, it applies counterfactual reasoning to generate outputs aligned with any desired value objectives. Benefitting from explicit causal modeling, COUPLE also provides better interpretability. We evaluate COUPLE on two datasets with different value systems and demonstrate that COUPLE advances other baselines across diverse types of value objectives.
Comment: This paper does not match any specific criteria but discusses counterfactual reasoning for value alignment in large language models, which might be of general interest. Relevance: 3 Novelty: 5
ArXiv ID: 2510.18229 Authors: Xinhao Cai, Liulei Li, Gensheng Pei, Tao Chen, Jinshan Pan, Yazhou Yao, Wenguan Wang
Abstract: This paper presents a generation-based debiasing framework for object detection. Prior debiasing methods are often limited by the representation diversity of samples, while naive generative augmentation often preserves the biases it aims to solve. Moreover, our analysis reveals that simply generating more data for rare classes is suboptimal due to two core issues: i) instance frequency is an incomplete proxy for the true data needs of a model, and ii) current layout-to-image synthesis lacks the fidelity and control to generate high-quality, complex scenes. To overcome this, we introduce the representation score (RS) to diagnose representational gaps beyond mere frequency, guiding the creation of new, unbiased layouts. To ensure high-quality synthesis, we replace ambiguous text prompts with a precise visual blueprint and employ a generative alignment strategy, which fosters communication between the detector and generator. Our method significantly narrows the performance gap for underrepresented object groups, \eg, improving large/rare instances by 4.4/3.6 mAP over the baseline, and surpassing prior L2I synthesis models by 15.9 mAP for layout accuracy in generated images.
Comment: This paper does not directly match any criteria but focuses on debiasing object detection using generative methods, which might be of general interest. Relevance: 3 Novelty: 5
ArXiv ID: 2510.18851 Authors: Rongyuan Wu, Lingchen Sun, Zhengqiang Zhang, Shihao Wang, Tianhe Wu, Qiaosi Yi, Shuai Li, Lei Zhang
Abstract: Benefiting from pre-trained text-to-image (T2I) diffusion models, real-world image super-resolution (Real-ISR) methods can synthesize rich and realistic details. However, due to the inherent stochasticity of T2I models, different noise inputs often lead to outputs with varying perceptual quality. Although this randomness is sometimes seen as a limitation, it also introduces a wider perceptual quality range, which can be exploited to improve Real-ISR performance. To this end, we introduce Direct Perceptual Preference Optimization for Real-ISR (DP$^2$O-SR), a framework that aligns generative models with perceptual preferences without requiring costly human annotations. We construct a hybrid reward signal by combining full-reference and no-reference image quality assessment (IQA) models trained on large-scale human preference datasets. This reward encourages both structural fidelity and natural appearance. To better utilize perceptual diversity, we move beyond the standard best-vs-worst selection and construct multiple preference pairs from outputs of the same model. Our analysis reveals that the optimal selection ratio depends on model capacity: smaller models benefit from broader coverage, while larger models respond better to stronger contrast in supervision. Furthermore, we propose hierarchical preference optimization, which adaptively weights training pairs based on intra-group reward gaps and inter-group diversity, enabling more efficient and stable learning. Extensive experiments across both diffusion- and flow-based T2I backbones demonstrate that DP$^2$O-SR significantly improves perceptual quality and generalizes well to real-world benchmarks.
Comment: This paper does not directly match any of the criteria but is related to generative modeling and image super-resolution, which might be of general interest. Relevance: 3 Novelty: 5
ArXiv ID: 2510.18633 Authors: Roxana Petcu, Kenton Murray, Daniel Khashabi, Evangelos Kanoulas, Maarten de Rijke, Dawn Lawrie, Kevin Duh
Abstract: Retrieval-augmented generation (RAG) systems address complex user requests by decomposing them into subqueries, retrieving potentially relevant documents for each, and then aggregating them to generate an answer. Efficiently selecting informative documents requires balancing a key trade-off: (i) retrieving broadly enough to capture all the relevant material, and (ii) limiting retrieval to avoid excessive noise and computational cost. We formulate query decomposition and document retrieval in an exploitation-exploration setting, where retrieving one document at a time builds a belief about the utility of a given sub-query and informs the decision to continue exploiting or exploring an alternative. We experiment with a variety of bandit learning methods and demonstrate their effectiveness in dynamically selecting the most informative sub-queries. Our main finding is that estimating document relevance using rank information and human judgments yields a 35% gain in document-level precision, 15% increase in {\alpha}-nDCG, and better performance on the downstream task of long-form generation.
Comment: Does not closely match any specific criterion but is generally relevant to retrieval-augmented generation and optimization techniques. Relevance: 3 Novelty: 5
ArXiv ID: 2510.18143 Authors: Huan Song, Deeksha Razdan, Yiyue Qian, Arijit Ghosh Chowdhury, Parth Patwa, Aman Chadha, Shinan Zhang, Sharlina Keshava, Hannah Marlowe
Abstract: Small Language Models (SLMs) offer compelling advantages in deployment cost and latency, but their accuracy often lags behind larger models, particularly for complex domain-specific tasks. While supervised fine-tuning can help bridge this performance gap, it requires substantial manual effort in data preparation and iterative optimization. We present PaDA-Agent (Pattern-guided Data Augmentation Agent), an evaluation-driven approach that streamlines the data augmentation process for SLMs through coordinated operations. Unlike state-of-the-art approaches that focus on model training errors only and generating error-correcting samples, PaDA-Agent discovers failure patterns from the validation data via evaluations and drafts targeted data augmentation strategies aiming to directly reduce the generalization gap. Our experimental results demonstrate significant improvements over state-of-the-art LLM-based data augmentation approaches for Llama 3.2 1B Instruct model fine-tuning.
Comment: This paper does not match any specific criteria but discusses data augmentation for fine-tuning small language models, which might be of general interest. Relevance: 3 Novelty: 4
ArXiv ID: 2510.18381 Authors: Giorgio Piras, Qi Zhao, Fabio Brau, Maura Pintor, Christian Wressnegger, Battista Biggio
Abstract: Adversarial pruning methods have emerged as a powerful tool for compressing neural networks while preserving robustness against adversarial attacks. These methods typically follow a three-step pipeline: (i) pretrain a robust model, (ii) select a binary mask for weight pruning, and (iii) finetune the pruned model. To select the binary mask, these methods minimize a robust loss by assigning an importance score to each weight, and then keep the weights with the highest scores. However, this score-space optimization can lead to sharp local minima in the robust loss landscape and, in turn, to an unstable mask selection, reducing the robustness of adversarial pruning methods. To overcome this issue, we propose a novel plug-in method for adversarial pruning, termed Score-space Sharpness-aware Adversarial Pruning (S2AP). Through our method, we introduce the concept of score-space sharpness minimization, which operates during the mask search by perturbing importance scores and minimizing the corresponding robust loss. Extensive experiments across various datasets, models, and sparsity levels demonstrate that S2AP effectively minimizes sharpness in score space, stabilizing the mask selection, and ultimately improving the robustness of adversarial pruning methods.
Comment: This paper does not match any specific criteria but discusses adversarial pruning, which is tangentially related to robustness in machine learning. Relevance: 3 Novelty: 4
ArXiv ID: 2510.18101 Authors: Vitor Pereira Matias, Daniel Perazzo, Vinicius Silva, Alberto Raposo, Luiz Velho, Afonso Paiva, Tiago Novello
Abstract: The problem of 3D reconstruction from posed images is undergoing a fundamental transformation, driven by continuous advances in 3D Gaussian Splatting (3DGS). By modeling scenes explicitly as collections of 3D Gaussians, 3DGS enables efficient rasterization through volumetric splatting, offering thus a seamless integration with common graphics pipelines. Despite its real-time rendering capabilities for novel view synthesis, 3DGS suffers from a high memory footprint, the tendency to bake lighting effects directly into its representation, and limited support for secondary-ray effects. This tutorial provides a concise yet comprehensive overview of the 3DGS pipeline, starting from its splatting formulation and then exploring the main efforts in addressing its limitations. Finally, we survey a range of applications that leverage 3DGS for surface reconstruction, avatar modeling, animation, and content generation-highlighting its efficient rendering and suitability for feed-forward pipelines.
Comment: This paper does not match any specific criteria but provides a tutorial on 3D Gaussian Splatting, which is related to 3D reconstruction and rendering. Relevance: 3 Novelty: 4
ArXiv ID: 2510.17940 Authors: Zhiming Lin
Abstract: Multi turn intent understanding is central to task oriented chatbots, yet real deployments face tight token budgets and noisy contexts, and most retrieval pipelines emphasize relevance while overlooking set level diversity and confounds such as more context or exemplar order. We ask whether retrieval diversity, rather than longer prompts, systematically improves LLM intent understanding under fixed budgets. We present a diversity aware retrieval framework that selects in context exemplars to balance intent coverage and linguistic variety, and integrates this selection with standard LLM decoders; the evaluation enforces budget matched prompts and randomized positions, and includes sensitivity analyses over exemplar count, diversity strength, and backbone size. On MultiWOZ 2.4 and SGD, the approach achieves strong gains in Joint Goal Accuracy under equal token budgets, surpassing strong LLM/DST baselines, with consistent improvements across K from 4 to 7 and moderate latency. Overall, the study isolates and validates the impact of content diversity in retrieval and offers a simple, deployable selection principle for building accurate, budget constrained multi turn intent systems.
Comment: This paper does not match any specific criteria but discusses retrieval diversity in multi-turn intent understanding, which is tangentially related to language models. Relevance: 3 Novelty: 4
In suggesting papers to your friend, remember that he enjoys papers on computer vision and machine learning, and generative modeling in multi-modal learning. Your friend also likes learning about surprising empirical or insightful results in vision-language models or embodied AI, as well as clever statistical tricks.