Current image watermarking methods are vulnerable to advanced image editing techniques enabled by large-scale text-to-image models. These models can distort embedded watermarks during editing, posing significant challenges to copyright protection. In this work, we introduce W-Bench, the first comprehensive benchmark designed to evaluate the robustness of watermarking methods against a wide range of image editing techniques, including image regeneration, global editing, local editing, and image-to-video generation. Through extensive evaluations of eleven representative watermarking methods against prevalent editing techniques, we demonstrate that most methods fail to detect watermarks after such edits. To address this limitation, we propose VINE, a watermarking method that significantly enhances robustness against various image editing techniques while maintaining high image quality. Our approach involves two key innovations: (1) we analyze the frequency characteristics of image editing and identify that blurring distortions exhibit similar frequency properties, which allows us to use them as surrogate attacks during training to bolster watermark robustness; (2) we leverage a large-scale pretrained diffusion model SDXL-Turbo, adapting it for the watermarking task to achieve more imperceptible and robust watermark embedding. Experimental results show that our method achieves outstanding watermarking performance under various image editing techniques, outperforming existing methods in both image quality and robustness. Code is available at https://github.com/Shilin-LU/VINE.
Medical Vision-Language Pretraining (MedVLP) shows promise in learning generalizable and transferable visual representations from paired and unpaired medical images and reports. MedVLP can provide useful features to downstream tasks and facilitate adapting task-specific models to new setups using fewer examples. However, existing MedVLP methods often differ in terms of datasets, preprocessing, and finetuning implementations. This pose great challenges in evaluating how well a MedVLP method generalizes to various clinically-relevant tasks due to the lack of unified, standardized, and comprehensive benchmark. To fill this gap, we propose BenchX, a unified benchmark framework that enables head-to-head comparison and systematical analysis between MedVLP methods using public chest X-ray datasets. Specifically, BenchX is composed of three components: 1) Comprehensive datasets covering nine datasets and four medical tasks; 2) Benchmark suites to standardize data preprocessing, train-test splits, and parameter selection; 3) Unified finetuning protocols that accommodate heterogeneous MedVLP methods for consistent task adaptation in classification, segmentation, and report generation, respectively. Utilizing BenchX, we establish baselines for nine state-of-the-art MedVLP methods and found that the performance of some early MedVLP methods can be enhanced to surpass more recent ones, prompting a revisiting of the developments and conclusions from prior works in MedVLP. Our code are available at https://github.com/yangzhou12/BenchX.
Transparent objects are ubiquitous in daily life, making their perception and robotics manipulation important. However, they present a major challenge due to their distinct refractive and reflective properties when it comes to accurately estimating the 6D pose. To solve this, we present ReFlow6D, a novel method for transparent object 6D pose estimation that harnesses the refractive-intermediate representation. Unlike conventional approaches, our method leverages a feature space impervious to changes in RGB image space and independent of depth information. Drawing inspiration from image matting, we model the deformation of the light path through transparent objects, yielding a unique object-specific intermediate representation guided by light refraction that is independent of the environment in which objects are observed. By integrating these intermediate features into the pose estimation network, we show that ReFlow6D achieves precise 6D pose estimation of transparent objects, using only RGB images as input. Our method further introduces a novel transparent object compositing loss, fostering the generation of superior refractive-intermediate features. Empirical evaluations show that our approach significantly outperforms state-of-the-art methods on TOD and Trans32K-6D datasets. Robot grasping experiments further demonstrate that ReFlow6D's pose estimation accuracy effectively translates to real-world robotics task. The source code is available at: https://github.com/StoicGilgamesh/ReFlow6D and https://github.com/StoicGilgamesh/matting_rendering.
Recently, sparsely-supervised 3D object detection has gained great attention, achieving performance close to fully-supervised 3D objectors while requiring only a few annotated instances. Nevertheless, these methods suffer challenges when accurate labels are extremely absent. In this paper, we propose a boosting strategy, termed SP3D, explicitly utilizing the cross-modal semantic prompts generated from Large Multimodal Models (LMMs) to boost the 3D detector with robust feature discrimination capability under sparse annotation settings. Specifically, we first develop a Confident Points Semantic Transfer (CPST) module that generates accurate cross-modal semantic prompts through boundary-constrained center cluster selection. Based on these accurate semantic prompts, which we treat as seed points, we introduce a Dynamic Cluster Pseudo-label Generation (DCPG) module to yield pseudo-supervision signals from the geometry shape of multi-scale neighbor points. Additionally, we design a Distribution Shape score (DS score) that chooses high-quality supervision signals for the initial training of the 3D detector. Experiments on the KITTI dataset and Waymo Open Dataset (WOD) have validated that SP3D can enhance the performance of sparsely supervised detectors by a large margin under meager labeling conditions. Moreover, we verified SP3D in the zero-shot setting, where its performance exceeded that of the state-of-the-art methods. The code is available at https://github.com/xmuqimingxia/SP3D.
The rise of large model-based AI agents has spurred interest in Multi-Agent Systems (MAS) for their capabilities in decision-making, collaboration, and adaptability. While the Model Context Protocol (MCP) addresses tool invocation and data exchange challenges via a unified protocol, it lacks support for organizing agent-level collaboration. To bridge this gap, we propose Agent-as-a-Service based on Agent Network (AaaS-AN), a service-oriented paradigm grounded in the Role-Goal-Process-Service (RGPS) standard. AaaS-AN unifies the entire agent lifecycle, including construction, integration, interoperability, and networked collaboration, through two core components: (1) a dynamic Agent Network, which models agents and agent groups as vertexes that self-organize within the network based on task and role dependencies; (2) service-oriented agents, incorporating service discovery, registration, and interoperability protocols. These are orchestrated by a Service Scheduler, which leverages an Execution Graph to enable distributed coordination, context tracking, and runtime task management. We validate AaaS-AN on mathematical reasoning and application-level code generation tasks, which outperforms state-of-the-art baselines. Notably, we constructed a MAS based on AaaS-AN containing agent groups, Robotic Process Automation (RPA) workflows, and MCP servers over 100 agent services. We also release a dataset containing 10,000 long-horizon multi-agent workflows to facilitate future research on long-chain collaboration in MAS.
Large Language Models (LLMs) have become increasingly central to recommendation scenarios due to their remarkable natural language understanding and generation capabilities. Although significant research has explored the use of LLMs for various recommendation tasks, little effort has been dedicated to verifying whether they have memorized public recommendation dataset as part of their training data. This is undesirable because memorization reduces the generalizability of research findings, as benchmarking on memorized datasets does not guarantee generalization to unseen datasets. Furthermore, memorization can amplify biases, for example, some popular items may be recommended more frequently than others. In this work, we investigate whether LLMs have memorized public recommendation datasets. Specifically, we examine two model families (GPT and Llama) across multiple sizes, focusing on one of the most widely used dataset in recommender systems: MovieLens-1M. First, we define dataset memorization as the extent to which item attributes, user profiles, and user-item interactions can be retrieved by prompting the LLMs. Second, we analyze the impact of memorization on recommendation performance. Lastly, we examine whether memorization varies across model families and model sizes. Our results reveal that all models exhibit some degree of memorization of MovieLens-1M, and that recommendation performance is related to the extent of memorization. We have made all the code publicly available at: https://github.com/sisinflab/LLM-MemoryInspector
This paper presents Diffusion via Autoregressive models (D-AR), a new paradigm recasting the image diffusion process as a vanilla autoregressive procedure in the standard next-token-prediction fashion. We start by designing the tokenizer that converts images into sequences of discrete tokens, where tokens in different positions can be decoded into different diffusion denoising steps in the pixel space. Thanks to the diffusion properties, these tokens naturally follow a coarse-to-fine order, which directly lends itself to autoregressive modeling. Therefore, we apply standard next-token prediction on these tokens, without modifying any underlying designs (either causal masks or training/inference strategies), and such sequential autoregressive token generation directly mirrors the diffusion procedure in image space. That is, once the autoregressive model generates an increment of tokens, we can directly decode these tokens into the corresponding diffusion denoising step in the streaming manner. Our pipeline naturally reveals several intriguing properties, for example, it supports consistent previews when generating only a subset of tokens and enables zero-shot layout-controlled synthesis. On the standard ImageNet benchmark, our method achieves 2.09 FID using a 775M Llama backbone with 256 discrete tokens. We hope our work can inspire future research on unified autoregressive architectures of visual synthesis, especially with large language models. Code and models will be available at https://github.com/showlab/D-AR
여러 에이전트(agent)가 동일한 저장소 내에서 동시에 쓰기 의도를 형성할 때, 실제 변경을 적용하기 전에 병렬 진행 가능 여부·직렬화 필요 여부·차단 여부를 결정하는 사전 쓰기 승인(pre-write admission) 문제를 다루며, ATM(AI-Atomic-Framework)은 CID 브로커와 어댑터 기반 원자화(atomization), 가상 원자(virtual atom) 폴백을 결합한 7단계 하드 게이트로 단일 거버넌스 도메인 내에서 감사 가능하고 결정론적인 승인 결정을 제공한다. 12개 시나리오 결정론적 설계 매트릭스와 ATM-AdmissionBench 등의 평가를 통해 단일 도메인 환경에서의 실현 가능성과 감사 가능성을 지원하나, Git 머지 대체나 크로스-클론 거버넌스는 범위 밖임을 명시한다.
의료 예약 스케줄링은 수동 조율, 분산된 레거시 시스템, 높은 행정 부담으로 인해 지속적인 운영상 병목 현상으로 남아 있다. 이러한 비효율성은 의료 제공자의 가용성을 제약하고 환자의 의료 접근성을 저하시킨다. 본 논문은 의료 물류 자동화를 위한 안전 우선 대화형 에이전트인 CareConnect를 제시한다. 이 에이전트는 대규모 언어 모델(LLM) 함수 호출, 검색-증강 생성(RAG), 계층화된 결정론적 안전 가드레일을 활용한다. 본 시스템은 예약 등록, 수정, 취소 및 시설 정보 검색을 지원하는 8개의 도메인 특화 도구를 조율하는 동시에, 의학적 조언이나 진단을 금지하는 엄격한 범위 제약을 적용한다. 안전 중요 상황은 응급 감지 및 의학적 의도 거부를 위한 결정론적 단락(short-circuit) 메커니즘을 통해 처리된다. 우리는 엔드투엔드 워크플로우, 멀티턴 상호작용, 엣지 케이스를 포괄하는 680개의 태스크 지향 시나리오로 구성된 종합 벤치마크에서 CareConnect를 평가한다. 실험 결과, 요청당 중앙값 지연 시간 2.2초에서 91.8%의 태스크 완료율, 전용 안전 중요 평가 서브셋에서 96.0%의 안전 준수율, 예약당 평균 운영 비용 $0.0324를 달성하여 수동 인간 스케줄링 대비 상당한 비용 절감을 실현하였다. 이러한 결과는 신중하게 범위가 설정되고 철저하게 안전장치가 마련된 LLM 기반 에이전트가 안전 보장을 유지하고 상당한 비용 효율성을 달성하면서 복잡한 의료 운영 워크플로우를 신뢰성 있게 자동화할 수 있음을 보여준다. 소스 코드와 시스템 구현은 https://github.com/Hadi-Hsn/CareConnect에서 공개적으로 이용 가능하다.
본 논문에서는 멀티모달 에이전트(multimodal agent)를 정적 이미지에서 연속적인 비디오 스트림으로 확장한 Video-DeepResearch(Video-DR)를 소개합니다. 이는 개방형 웹 탐험(exploration)과 결합된 조밀한 시공간적 그라운딩(spatiotemporal grounding)을 요구하는 설정입니다. 사전 평가 결과 현재 모델들에서 두 가지 치명적인 병목 현상이 드러났습니다. 첫째는 에이전트가 시각적 도구를 건너뛰고 텍스트 검색을 선호하는 모달리티 바이어스(modality bias)이며, 둘째는 모델이 진정한 도구 증강 실행 대신 내부 메모리에 의존하는 파라메트릭 지식 누출(parametric knowledge leakage)입니다. 이러한 문제를 해결하기 위해, 웹 검색에 앞서 프레임 간 철저한 시각적 그라운딩을 강제하는 단계별 도구 잠금 해제 기능을 갖춘 분리형 인지-탐험 파이프라인의 Video-DR을 제안합니다. 본 프레임워크는 지도 미세조정(supervised fine-tuning) 후 그룹 상대 정책 최적화(GRPO, Group Relative Policy Optimization)를 수행하는 2단계 학습(training) 레시피를 채택하여, 모방 학습의 한계를 극복하는 자율적 탐험을 가능하게 합니다. 나아가 200개의 복잡한 다단계(multi-hop) VQA 인스턴스로 구성된 인간-AI 협업 벤치마크인 Video-DR-Bench를 구축했습니다. 실증적 결과에 따르면 Video-DeepResearch-35B-A3B는 평균 정확도 64.0%로 새로운 최고 성능(SOTA)을 달성하여 상용 모델인 Claude-4.5-Sonnet(59.0%)을 5.0% 포인트 앞서고, GPT-5(52.5%) 및 Gemini 2.5 Pro(57.5%)를 크게 능가했습니다. 30B-A3B 변형 모델은 59.3%를 달성하여 Claude-4.5-Sonnet과 대등한 경쟁력을 보였으며, 컴팩트한 규모에서도 본 학습 패러다임의 효과성을 입증했습니다. 코드는 https://github.com/Osilly/Vision-DeepResearch 에서 확인할 수 있습니다.
에이전트는 결정적인 런타임 결과가 관찰되기 전에 실행 구조를 최적화·선택·제약하는 경향이 있다. 그러나 이러한 실행 전 확정은 오케스트레이션 병목을 만든다. 중간 증거가 대기 중인 후속 실행을 무효화하면, 에이전트는 낡은 단계를 실행하거나 광범위하게 재계획해야 하며, 오류가 누적되고 계산이 낭비되며 진행이 버려진다. 따라서 우리는 Trace-grounded Route Orchestration via Validation and Editing(TROVE)를 제안한다. TROVE는 런타임 증거가 무효화하는 부분만 수정한다. 오프라인에서 TROVE는 평가된 워크플로우 검색 trace를 원자적·복합 스킬과 결과 조건부 전환 그래프로 증류하여, 안정적인 조각은 보존하면서 결과 의존적 결정을 드러낸다. 온라인에서는 계획된 경로를 잠정적인 것으로 보며, 최상위 스킬 하나를 확정한 뒤 컨트롤러는 유효한 후속을 유지하거나 trace가 뒷받침하는 국소 대응을 삽입하거나 무효 접미사만 교체한다. 서로 다른 LLM 백본을 사용한 코드 생성, 질의 응답, 수학 추론 벤치마크에 대한 평가 결과, TROVE는 데이터셋 수준 최적화, 쿼리 수준 아키텍처 선택, 그래프 제약 스케줄링 등 기존 baseline보다 더 나은 품질-효율 트레이드오프를 제공한다. 품질 향상은 결과가 적절한 후속 실행을 바꿀 때 가장 크며, 조기 종료는 거의 포화에 가까운 태스크에서 상당한 효율 이득을 준다. Ablation은 복합 스킬이 오프라인 이점의 대부분을 포착하고, 삽입이 국소 수정을 가능하게 하며, 접미사 교체가 주로 효율을 개선함을 추가로 보여준다. 이러한 결과는 선택적 경로 편집이 적응형 에이전트 오케스트레이션의 일반 원칙임을 보여준다.
온-정책 증류(On-Policy Distillation, OPD)는 교사 모델의 피드백을 활용하여 학생 모델이 자신의 생성 결과로부터 학습하도록 한다. 토크나이저가 서로 다른 경우, 교사와 학생의 예측을 비교하려면 시퀀스 수준과 어휘(vocabulary) 수준 모두에서 정렬이 필요하다. 본 논문에서는 이러한 정렬 커버리지를 확장하는 것이 학습(training)을 개선하는지 검토한다. 수학적 추론과 코드 생성 과제에 대한 세 가지 이질적인 교사-학생 모델 쌍에서, 엄격한 1:1 그룹은 상당한 어휘 불일치에도 불구하고 이미 학생이 생성한 토큰의 대부분을 커버한다. 증류 이전에 학생 모델로부터 샘플링한 응답에서, 공유 어휘(shared vocabulary)는 평균적으로 엄격하게 정렬된 위치에서 교사와 학생의 확률 질량(probability mass)을 거의 전부 보존한다. 각 엄격한 위치에서 역방향 KL(reverse KL)을 학생이 선택한 공유 어휘의 상위 16개 부분집합으로 제한하면 전체 공유 어휘를 사용한 OPD와 비슷한 정확도를 달성하며, 평가된 교차 토크나이저 베이스라인들을 능가한다. 불일치 그룹의 스팬(span) 로그 확률에 평균제곱오차(MSE) 감독을 추가하면 완전한 감독 커버리지를 얻을 수 있지만, 오히려 정확도는 감소한다. 엄격한 손실(loss)만으로 학습한 체크포인트에서, 스팬 그래디언트는 엄격한 그래디언트와 약하거나 음의 방향 일치도를 보이며, 그 크기는 엄격한 그래디언트에 비해 점점 커진다. 이러한 진단 결과는 스팬 감독을 추가함으로써 발생하는 정확도 저하를 설명하는 데 도움이 될 수 있다. 본 연구의 결과는 정렬 커버리지를 최대화하는 접근에서 감독 신뢰성을 우선시하는 접근으로의 전환을 시사한다: 엄격한 위치에서의 압축된 감독이, 약하게 정렬되거나 상충하는 학습 신호를 유발하는 더 넓은 커버리지보다 더 효과적일 수 있다.
Abliteration, a fine-tuning-free weight edit, removes refusal from open-weight large language models (LLMs). The developer of Arbitrary-Rank Ablation (ARA), an optimisation-based variant, reports that direction-based abliteration, which ablates a refusal direction, increasingly fails on newer models. On five release pairs, refusal removal did not become uniformly harder. Qwen2.5-32B and Qwen3.8-27B, Gemma-2-27B and Gemma-4-31B, OLMo-2-32B and Olmo-3.1-32B, Mistral-Small-2409 and Mistral-Small-3.2, and Qwen2.5-7B and Qwen3.5-4B were edited in Heretic 2.0 by three or four direction-based methods (difference-in-means, MPOA, SOM and, on the small pair, J-space) and ARA under the same prompts, nominal search budget, first-token KL divergence cap and selection rule; removal counted as harder when the automatic WildGuard judge found more of 972 held-out harmful prompts (JailbreakBench, XSTest, StrongREJECT, HarmBench and an author-compiled malicious-code bank) still refused. After direction-based removal, newer releases retained -1 to +37 percentage points more refusal, and after ARA -26 to +34. With run-to-run variance included, between-pair heterogeneity of the direction-based effect ranged from high to low with the scale and variance model (I^2 20-80%). ARA lost less effectiveness than direction-based methods in Gemma and OLMo and more in Qwen; each verdict rests on one ARA run, and some changed with the divergence budget or variance model. In Gemma the direction-based increase came from the malicious-code bank and reversed on public sets. Contrary to prediction, keyword counts at the tool's window did not undercount judged refusal on average, but differed by a mean 7 points per model. Audits of a single pair, method or prompt source can reach opposite conclusions. Whether a newer model resisted removal depended on the release pair, method and budget, not on generation alone.
The main contribution of this paper is a simple semi-supervised pipeline that only uses the original training set without collecting extra data. It is challenging in 1) how to obtain more training data only from the training set and 2) how to use the newly generated data. In this work, the generative adversarial network (GAN) is used to generate unlabeled samples. We propose the label smoothing regularization for outliers (LSRO). This method assigns a uniform label distribution to the unlabeled images, which regularizes the supervised model and improves the baseline. We verify the proposed method on a practical problem: person re-identification (re-ID). This task aims to retrieve a query person from other cameras. We adopt the deep convolutional generative adversarial network (DCGAN) for sample generation, and a baseline convolutional neural network (CNN) for representation learning. Experiments show that adding the GAN-generated data effectively improves the discriminative ability of learned CNN embeddings. On three large-scale datasets, Market-1501, CUHK03 and DukeMTMC-reID, we obtain +4.37%, +1.6% and +2.46% improvement in rank-1 precision over the baseline CNN, respectively. We additionally apply the proposed method to fine-grained bird recognition and achieve a +0.6% improvement over a strong baseline. The code is available at https://github.com/layumi/Person-reID_GAN.
Symbolic regression (SR) is a data analysis problem where we search for the mathematical expression that best fits a numerical dataset. It is a global optimization problem. The most popular approach to SR is by genetic programming (SRGP). It is a common paradigm to compare an algorithm's performance to that of random search, but the data comparing SRGP to random search is lacking. We describe a novel algorithm for SR, namely SR by uniform random global search (SRURGS), also known as pure random search. We conduct experiments comparing SRURGS with SRGP using 100 randomly generated equations. Our results suggest that a SRGP is faster than SRURGS in producing equations with good R^2 for simple problems. However, our experiments suggest that SRURGS is more robust than SRGP, able to produce good output in more challenging problems. As SRURGS is arguably the simplest global search algorithm, we believe it should serve as a control algorithm against which other symbolic regression algorithms are compared. SRURGS has only one tuning parameter, and is conceptually very simple, making it a useful tool in solving SR problems. The method produces random equations, which is useful for the generation of symbolic regression benchmark problems. We have released well documented and open-source python code, currently under formal peer-review, so that interested researchers can deploy the tool in practice.
It has been a long time that computer architecture and systems are optimized for efficient execution of machine learning (ML) models. Now, it is time to reconsider the relationship between ML and systems, and let ML transform the way that computer architecture and systems are designed. This embraces a twofold meaning: improvement of designers' productivity, and completion of the virtuous cycle. In this paper, we present a comprehensive review of the work that applies ML for computer architecture and system design. First, we perform a high-level taxonomy by considering the typical role that ML techniques take in architecture/system design, i.e., either for fast predictive modeling or as the design methodology. Then, we summarize the common problems in computer architecture/system design that can be solved by ML techniques, and the typical ML techniques employed to resolve each of them. In addition to emphasis on computer architecture in a narrow sense, we adopt the concept that data centers can be recognized as warehouse-scale computers; sketchy discussions are provided in adjacent computer systems, such as code generation and compiler; we also give attention to how ML techniques can aid and transform design automation. We further provide a future vision of opportunities and potential directions, and envision that applying ML for computer architecture and systems would thrive in the community.
Point normal, as an intrinsic geometric property of 3D objects, not only serves conventional geometric tasks such as surface consolidation and reconstruction, but also facilitates cutting-edge learning-based techniques for shape analysis and generation. In this paper, we propose a normal refinement network, called Refine-Net, to predict accurate normals for noisy point clouds. Traditional normal estimation wisdom heavily depends on priors such as surface shapes or noise distributions, while learning-based solutions settle for single types of hand-crafted features. Differently, our network is designed to refine the initial normal of each point by extracting additional information from multiple feature representations. To this end, several feature modules are developed and incorporated into Refine-Net by a novel connection module. Besides the overall network architecture of Refine-Net, we propose a new multi-scale fitting patch selection scheme for the initial normal estimation, by absorbing geometry domain knowledge. Also, Refine-Net is a generic normal estimation framework: 1) point normals obtained from other methods can be further refined, and 2) any feature module related to the surface geometric structures can be potentially integrated into the framework. Qualitative and quantitative evaluations demonstrate the clear superiority of Refine-Net over the state-of-the-arts on both synthetic and real-scanned datasets. Our code is available at https://github.com/hrzhou2/refinenet.
Aspect-based sentiment analysis (ABSA) is a fine-grained sentiment analysis task, which focuses on detecting the sentiment polarity towards the aspect in a sentence. However, it is always sensitive to the multi-aspect challenge, where features of multiple aspects in a sentence will affect each other. To mitigate this issue, we design a novel training framework, called Contrastive Cross-Channel Data Augmentation (C3 DA), which leverages an in-domain generator to construct more multi-aspect samples and then boosts the robustness of ABSA models via contrastive learning on these generated data. In practice, given a generative pretrained language model and some limited ABSA labeled data, we first employ some parameter-efficient approaches to perform the in-domain fine-tuning. Then, the obtained in-domain generator is used to generate the synthetic sentences from two channels, i.e., Aspect Augmentation Channel and Polarity Augmentation Channel, which generate the sentence condition on a given aspect and polarity respectively. Specifically, our C3 DA performs the sentence generation in a cross-channel manner to obtain more sentences, and proposes an Entropy-Minimization Filter to filter low-quality generated samples. Extensive experiments show that our C3 DA can outperform those baselines without any augmentations by about 1% on accuracy and Macro- F1. Code and data are released in https://github.com/wangbing1416/C3DA.
Detecting which nodes in graphs are outliers is a relatively new machine learning task with numerous applications. Despite the proliferation of algorithms developed in recent years for this task, there has been no standard comprehensive setting for performance evaluation. Consequently, it has been difficult to understand which methods work well and when under a broad range of settings. To bridge this gap, we present--to the best of our knowledge--the first comprehensive benchmark for unsupervised outlier node detection on static attributed graphs called BOND, with the following highlights. (1) We benchmark the outlier detection performance of 14 methods ranging from classical matrix factorization to the latest graph neural networks. (2) Using nine real datasets, our benchmark assesses how the different detection methods respond to two major types of synthetic outliers and separately to "organic" (real non-synthetic) outliers. (3) Using an existing random graph generation technique, we produce a family of synthetically generated datasets of different graph sizes that enable us to compare the running time and memory usage of the different outlier detection algorithms. Based on our experimental results, we discuss the pros and cons of existing graph outlier detection algorithms, and we highlight opportunities for future research. Importantly, our code is freely available and meant to be easily extendable: https://github.com/pygod-team/pygod/tree/main/benchmark
Machine learning recently proved efficient in learning differential equations and dynamical systems from data. However, the data is commonly assumed to originate from a single never-changing system. In contrast, when modeling real-world dynamical processes, the data distribution often shifts due to changes in the underlying system dynamics. Continual learning of these processes aims to rapidly adapt to abrupt system changes without forgetting previous dynamical regimes. This work proposes an approach to continual learning based on reservoir computing, a state-of-the-art method for training recurrent neural networks on complex spatiotemporal dynamical systems. Reservoir computing fixes the recurrent network weights - hence these cannot be forgotten - and only updates linear projection heads to the output. We propose to train multiple competitive prediction heads concurrently. Inspired by neuroscience's predictive coding, only the most predictive heads activate, laterally inhibiting and thus protecting the inactive heads from forgetting induced by interfering parameter updates. We show that this multi-head reservoir minimizes interference and catastrophic forgetting on several dynamical systems, including the Van-der-Pol oscillator, the chaotic Lorenz attractor, and the high-dimensional Lorenz-96 weather model. Our results suggest that reservoir computing is a promising candidate framework for the continual learning of dynamical systems. We provide our code for data generation, method, and comparisons at \url{https://github.com/leonardbereska/multiheadreservoir}.