Singapore

Visual Place Recognition (VPR) aims to determine the geographic location of a query image by retrieving its most visually similar counterpart from a geo-tagged reference database. Recently, the emergence of the powerful visual foundation model, DINOv2, trained in a self-supervised manner on massive datasets, has significantly improved VPR performance. This improvement stems from DINOv2’s exceptional feature generalization capabilities but is often accompanied by increased model complexity and computational overhead that impede deployment on resource-constrained devices. To address this challenge, we propose $D^{2}$-VPR, a $D$istillation- and $D$eformable-based framework that retains the strong feature extraction capabilities of visual foundation models while significantly reducing model parameters and achieving a more favorable performance-efficiency trade-off. Specifically, first, we employ a two-stage training strategy that integrates knowledge distillation and fine-tuning. Additionally, we introduce a Distillation Recovery Module (DRM) to better align the feature spaces between the teacher and student models, thereby minimizing knowledge transfer losses to the greatest extent possible. Second, we design a Top-Down-attention-based Deformable Aggregator (TDDA) that leverages global semantic features to dynamically and adaptively adjust the Regions of Interest (ROI) used for aggregation, thereby improving adaptability to irregular structures. Extensive experiments demonstrate that our method achieves competitive and even superior performance compared to state-of-the-art approaches. Meanwhile, it reduces the parameter count by approximately 64.2\% and FLOPs by about 62.6\% (compared to CricaVPR).

AAAI 2026

D²-VPR: A Parameter-efficient Visual-foundation-model-based Visual Place Recognition Method via Knowledge Distillation and Deformable Aggregation

visual place recognition

parameter-efficient

image retrieval

We are pleased to announce the Thirty-Eighth AAAI Conference on Artificial Intelligence (AAAI-24), which will be held in Vancouver, British Columbia at the Vancouver Convention Centre – West Building from 20-27 February, 2024.

The purpose of the AAAI conference series is to promote research in Artificial Intelligence (AI) and foster scientific exchange between researchers, practitioners, scientists, students, and engineers across the entirety of AI and its affiliated disciplines. AAAI-24 will feature technical paper presentations, special tracks, invited speakers, workshops, tutorials, poster sessions, senior member presentations, competitions, and exhibit programs, and a range of other activities to be announced.

We expect for AAAI-24 to be an in-person conference – one author of all accepted papers will be expected to present work in person unless there are exceptional circumstances that prevent this.

In order to access the AAAI-24 event page you need to register [here](https://aaai.org/aaai-conference/registration/)

AAAI 2024

The purpose of the AAAI conference series is to promote research in Artificial Intelligence (AI) and foster scientific exchange between researchers, practitioners, scientists, students, and engineers across the entirety of AI and its affiliated disciplines. 

technical paper

We are pleased to announce the Fortieth AAAI Conference on Artificial Intelligence (AAAI-26), which will be held in Singapore EXPO from January 20 to January 27, 2026.

The purpose of the AAAI conference series is to promote research in Artificial Intelligence (AI) and foster scientific exchange between researchers, practitioners, scientists, students, and engineers across the entirety of AI and its affiliated disciplines. AAAI-26 will feature technical paper presentations, special tracks, invited speakers, workshops, tutorials, poster sessions, senior member presentations, competitions, and exhibit programs, and a range of other activities to be announced.

To access this event page, you need to log in with the **email address you registered with**. Access credentials will be sent to your email from Underline - subject line "Welcome to AAAI 2026". Please be sure to check your spam email folder if you do not see an email confirmation right away.

Please log in

To access this event page, you are required to register.
Please complete your registration to continue.

We recommend reading [**the registration information**](https://aaai.org/conference/aaai/aaai-26/registration/) first.

**Online Registration Form**: https://aaai.getregistered.net/conference-2026 

Registration Required

We are pleased to announce the Fortieth AAAI Conference on Artificial Intelligence (AAAI-26), which will be held in Singapore EXPO from January 20 to January 27, 2026.

This work presents Insert Anything, a unified framework for reference-based image insertion that seamlessly integrates objects from reference images into target scenes under flexible, user-specified control guidance. Instead of training separate models for individual tasks, our approach is trained once on our new AnyInsertion dataset, the first open-source large-scale dataset specifically designed for reference image–based image editing, comprising 120K prompt-image pairs covering diverse tasks such as person, object, and garment insertion--and effortlessly generalizes to a wide range of insertion scenarios. Such a challenging setting requires capturing both identity features and fine-grained details, while allowing versatile local adaptations in style, color, and texture. To this end, we propose to leverage the multimodal attention of the Diffusion Transformer (DiT) to support both mask- and text-guided editing. Furthermore, we introduce an in-context editing mechanism that treats the reference image as contextual information, employing two prompting strategies to harmonize the inserted elements with the target scene while faithfully preserving their distinctive features. Extensive experiments on AnyInsertion, DreamBooth, and VTON-HD benchmarks demonstrate that our method consistently outperforms existing alternatives, underscoring its great potential in real-world applications such as creative content generation, virtual try-on, and scene composition.

Insert Anything: Image Insertion via In-Context Editing in DiT

As automated attack techniques rapidly advance, CAPTCHAs remain a critical defense mechanism against malicious bots. However, existing CAPTCHA
schemes encompass a diverse range of modalities—from static distorted text and
obfuscated images to interactive clicks, sliding puzzles, and logic-based questions—yet the community still lacks a unified, large-scale, multimodal benchmark to rigorously evaluate their security robustness. To address this gap, we
introduce MCA-Bench, a comprehensive and reproducible benchmarking suite
that integrates heterogeneous CAPTCHA types into a single evaluation protocol.
Leveraging a shared vision–language model backbone, we fine-tune specialized
cracking agents for each CAPTCHA category, enabling consistent, cross-modal
assessments. Extensive experiments reveal that MCA-Bench effectively maps
the vulnerability spectrum of modern CAPTCHA designs under varied attack
settings, and—crucially—offers the first quantitative analysis of how challenge
complexity, interaction depth, and model solvability interrelate. Based on these
findings, we propose three actionable design principles and identify key open
challenges, laying the groundwork for systematic CAPTCHA hardening, fair
benchmarking, and broader community collaboration.

MCA-Bench: A Multimodal Benchmark for Evaluating CAPTCHA Robustness Against VLM-based Attacks

RGB-to-RAW reconstruction, or the reverse modeling of a camera Image Signal Processing (ISP) pipeline, aims to recover high-fidelity RAW data from RGB images. Despite notable progress, existing learning-based methods typically treat this task as a direct regression objective and still struggle with detail inconsistency and color deviation, due to the ill-posed nature of inverse ISP and the inherent information loss in quantized RGB images. To address these limitations, we pioneer a generative perspective by reformulating RGB-to-RAW reconstruction as a deterministic latent transport problem and introduce a novel framework named $\textbf{RAW-Flow}$, which leverages flow matching to learn a deterministic vector field in latent space, to effectively bridge the gap between RGB and RAW representations and enable accurate reconstruction of structural details and color information. To further enhance latent transport, we introduce a cross-scale context guidance module that injects hierarchical RGB features into the flow estimation process. Moreover, we design a Dual-domain Latent Autoencoder (DLAE) with a feature alignment constraint to support the proposed latent transport framework, which jointly encodes RGB and RAW inputs while promoting stable training and high-fidelity reconstruction. Extensive experiments demonstrate that RAW-Flow outperforms state-of-the-art approaches both quantitatively and visually. Code will be released to facilitate future research.

RAW-Flow: Advancing RGB-to-RAW Image Reconstruction with Deterministic Latent Flow Matching

In recommender systems, recent advances highlight the critical role of alignment and uniformity (AU) in representation learning. Specifically, AU-based methods pull positive user-item pairs closer (alignment) and spread the overall representation distribution (uniformity), typically relying on observed positive samples. Despite their effectiveness, exist methods face two limitations: (1) noise issues have a more severe impact on AU-based methods in the absence of negative samples, leading to the capture of spurious signals such as misclicks or non-preferential behaviors; (2) data sparsity weakens the alignment of user-item representations, hindering reliable representation learning and harming recommendations for sparse users. To tackle these issues, we propose a novel recommendation framework named Constrained Alignment and Filtered Uniformity (CAFU). CAFU enhances robustness through Filtered Uniformity (FU) and improves performance under data sparsity via Constrained Alignment (CA). Specifically, FU adopts a threshold-based strategy to eliminate unreliable samples that degrade embedding quality, thereby strengthening robustness. In parallel, CA mitigates the impact of sparsity by masking low-confidence user-item pairs based on angular distance, leading to better recommendation for sparse users. Extensive experiments on three datasets and three backbones demonstrate the effectiveness and generalization of the proposed framework.

CAFU: Constrained Alignment and Filtered Uniformity for Denoising Recommendation

Neuroscientific evidence reveals that human visual recognition is not an instantaneous event but a hierarchical process, where the brain constructs a holistic perception by progressively integrating simple features like edges or texture into complex scenes. Ensemble learning successfully operationalizes this principle, yet existing methods typically integrate models at the decision level, neglecting the rich, complementary information within the feature space itself and thus fundamentally limiting their potential. To address this, we introduce Synergistic Semantic Boosting (S²-Boosting), a framework that first employs a self-supervised hierarchical semantic learning module to decompose an image into complementary, semantically meaningful parts autonomously. These parts guide a boosting procedure where a sequence of specialized learners, each focusing on a specific partition, collaboratively corrects the ensemble's errors. We further present encouraging results on real-world image datasets, highlighting the intrinsic interpretability, paving the way for more robust and transparent models. The code will be available at https://anonymous.4open.science/r/S-2Boosting-0D47/.

S2-Boost: Synergistic Semantic Boosting for Coarse-to-Fine Ensemble Learning

Higher-order Graph Neural Networks (HOGNNs), particularly those based on the 2-Folklore Weisfeiler-Leman (2-FWL) framework, achieve superior expressive power by modeling 2- and 3-node interactions, enabling fine-grained capture of graph structure beyond standard GNNs. However, their $\mathcal{O}(n^3)$ computational and memory complexity limits scalability, prompting efficiency-focused methods that often sacrifice expressivity. We propose **Co-Sparsify**, a connectivity-aware sparsification framework that removes only computation redundant for expressivity, preserving full 2-FWL discriminative power while significantly improving efficiency. Our key insight is that 3-node interactions are necessary only within *biconnected components*---maximal subgraphs without cut nodes, where any two nodes are connected by at least two internally disjoint paths. Within such components, joint modeling of intermediate nodes provides additional expressive power. In contrast, when node pairs are connected only through a cut node or belong to different connected components, their structural relationships can be decomposed into 2-node interactions or captured globally via readout, rendering higher-order modeling redundant. By restricting 2-node message passing to connected components and 3-node interactions to biconnected components---using an efficient $O(n + m)$ block-cut tree decomposition---Co-Sparsify eliminates provably redundant computations without approximation or sampling. We prove that Co-Sparsified GNNs are as powerful as the 2-FWL test. Empirical evaluation on PPGN shows preserved or improved accuracy on synthetic substructure counting tasks and state-of-the-art performance on real-world graph benchmarks such as ZINC and QM9. This work demonstrates that high expressivity and efficiency can coexist through principled, topology-guided sparsification, offering a scalable path for powerful GNNs without compromising theoretical guarantees.

Connectivity-Guided Sparsification of 2-FWL GNNs: Preserving Full Expressivity with Improved Efficiency

The advancement of intelligent agents has revolutionized problem-solving across diverse domains, yet solutions for personalized fashion styling remain underexplored, which holds immense promise for promoting shopping experiences. In this work, we present StyleTailor, the first collaborative agent framework that seamlessly unifies personalized apparel design, shopping recommendation, virtual try-on, and systematic evaluation into a cohesive workflow. To this end, StyleTailor pioneers an iterative visual refinement paradigm driven by _multi-level negative feedback_, enabling adaptive and precise user alignment. Specifically, our framework features two core agents, _i.e.,_ _Designer_ for personalized garment selection and _Consultant_ for virtual try-on, whose outputs are progressively refined via hierarchical vision-language model feedback spanning individual items, complete outfits, and try-on efficacy. Counterexamples are aggregated into negative prompts, forming a closed-loop mechanism that enhances recommendation quality.To assess the performance, we introduce a _comprehensive evaluation suite_ encompassing style consistency, visual quality, face similarity, and artistic appraisal. Extensive experiments demonstrate StyleTailor's superior performance in delivering personalized designs and recommendations, outperforming strong baselines without negative feedback and establishing a new benchmark for intelligent fashion systems.

StyleTailor: Towards Personalized Fashion Styling via Hierarchical Negative Feedback

Teleoperation can enable human intervention to help handle
instances of failure in autonomy thus allowing for much
safer deployment of autonomous vehicle technology.
Successful teleoperation requires recreating the
environment around the remote vehicle using camera data
received over wireless communication channels. This paper
develops a new predictive display system to tackle the
significant time delays encountered in receiving camera
data over wireless networks. First, a new high gain
observer is developed for estimating the position and
orientation of the ego vehicle. The novel observer is shown
to perform accurate state estimation using only GNSS and
gyroscope sensor readings. A vector field method which
fuses the delayed camera and Lidar data is then presented.
This method uses sparse 3D points obtained from Lidar and
transforms them using the state estimates from the high
gain observer to generate a sparse vector field for the
camera image. Polynomial based interpolation is then
performed to obtain the vector field for the complete image
which is then remapped to synthesize images for accurate
predictive display. The method is evaluated on real-world
experimental data from the nuScenes and KITTI datasets. The
performance of the high gain observer is also evaluated and
compared with that of the EKF. The synthesized images using
the vector field based predictive display are compared with
ground truth images using various image metrics and offer
vastly improved performance compared to delayed images.

Predictive Display for Teleoperation Based on Vector Fields
Using Lidar-Camera Fusion

Large Language Models (LLMs) have rapidly advanced and are widely adopted across diverse fields. Due to the substantial computational cost and data requirements of training from scratch, many developers choose to fine-tune or modify existing open-source models. While most adhere to open-source licenses, some falsely claim original training despite clear derivation from public models. This raises pressing concerns about intellectual property protection and highlights the need for reliable methods to verify model provenance.
In this paper, we propose GhostSpec, a lightweight yet effective method for verifying LLM lineage without access to training data or modification of model behavior. Our approach constructs compact and robust fingerprints by applying singular value decomposition (SVD) to invariant products of internal attention weight matrices, effectively capturing the structural identity of a model. Unlike watermarking or output-based methods, GhostSpec is fully data-free, non-invasive, and computationally efficient. It demonstrates strong robustness to sequential fine-tuning, pruning, block expansion, and even adversarial transformations.
Extensive experiments show that GhostSpec can reliably trace the lineage of transformed models with minimal overhead. By offering a practical solution for model verification and reuse tracking, our method contributes to the protection of intellectual property and fosters a transparent, trustworthy ecosystem for large-scale language models.

Ghost in the Transformer: Detecting Model Reuse with Invariant Spectral Signatures

Recent advancements in Text-to-Speech (TTS) technology have been remarkable, enabling current models to clone arbitrary unseen speakers and synthesize high-quality, natural-sounding speech. However, corresponding evaluation techniques appear to be lagging: Existing Mean Opinion Score (MOS) estimation models typically perform regression-based scoring on entire speech segments—while a failed synthesized speech usually contains problematic elements in only a few isolated words rather than throughout the entire utterance. In this context, we present an intriguing finding: encoder-decoder ASR models, such as Whisper, leverage their extensive pre-training to precisely capture word-level mismatches between speech and text within their cross-attention mechanisms, thereby providing a fine-grained reward signal. Building upon this insight, we propose a novel TTS optimization method, which we term Word-level TTS Alignment by ASR-driven Attentive Reward (W3AR). Instead of relying on any explicit reward annotations, W3AR leverages the attention information within a pre-trained ASR model, enabling finer-grained alignment and optimization of the sequences predicted by the TTS model. Experimental results demonstrate that W3AR not only effectively improves the TTS generation quality of existing models but also further enhances zero-shot robustness on unseen speakers. Additionally, our findings and proposed methodology offer a new insight for generative tasks: understanding models can potentially serve as evaluators, providing highly fine-grained and valuable feedback for generative optimization.

Downloads

Next from AAAI 2026

Insert Anything: Image Insertion via In-Context Editing in DiT

Similar lecture

DistilVPR: Cross-Modal Knowledge Distillation for Visual Place Recognition | VIDEO