GRC‑Net is a new unsupervised multimodal anomaly detection framework that uses a global‑attention MLP to enforce consistency across patch embeddings and a stable reconstruction module to reduce noise. By capturing holistic context through a global token, it minimizes discrepancies between original and predicted embeddings, leading to more reliable reconstruction errors. Experiments on MVTec 3D‑AD and Eyecandies show that GRC‑Net outperforms existing methods at both image and pixel levels, with clearer distinctions between normal and anomalous regions.
By Seyoung Jeong, Jong Pil Yun, Sang Jun Lee
The paper introduces temporally interpretable differentiable decision trees (DDTs) that incorporate action chunking to better align tree-based decision making with multi‑timestep planning. Two new policy‑gradient algorithms and an information‑theoretic tree‑restructuring method are proposed to enable efficient training of these trees. Experiments on four simulation environments show that warm‑starting action‑chunked DDTs from a distilled policy yields trees that match neural network performance in three domains while using up to 80% fewer parameters.
By Eisuke Hirota, Aarav Sane, Rohan Paleja
OrBIT is a structure‑guided embedding compression framework that learns reusable local geometry from orbit dynamics to constrain a small set of shared codewords. It allocates a fixed coding budget based on a global reconstruction residual, allowing redundant overlapping charts to compensate local errors after gluing. Across four large language model embedding tables, OrBIT achieves 37.9× compression on GPT‑2 and over 23× on each 7B table compared to 16‑bit storage, while maintaining competitive rate‑distortion performance against existing quantization and low‑rank methods.
By Yunied Puig, Amit Kumar Jaiswal
The paper introduces an extended deep energy method for modeling thermo‑mechanical crack propagation, keeping the crack as a sharp polyline. Two neural networks predict temperature and displacement, receiving the crack via a scalar embedding that allows discontinuities without regularization, and the displacement is enriched near the tip using a trainable Williams expansion. The method minimizes incremental conduction and thermoelastic energy, employs Monte Carlo integration, and extracts stress intensity factors through an interaction integral, validating results against published benchmarks for thermal edge cracks, graded shear tests, and notched cruciform specimens.
By Han Zhang, Mehrisadat Makki Alamdari, Babak Shahbodagh, Mohammad Vahab, Cosmin Anitescu, Timon Rabczuk, Elena Atroshchenko
PrismSSL is a Python library that consolidates leading self‑supervised learning methods for audio, vision, graphs, and cross‑modal tasks into a single, modular codebase. It allows users to install, configure, and run pretext training with minimal code, reproduce compact benchmarks, and extend the framework with new modalities or methods via clean trainer and dataset abstractions. The library, available on PyPI under an MIT license, integrates with HuggingFace Transformers and offers features such as distributed PyTorch training, Optuna hyperparameter search, LoRA fine‑tuning, animated embedding visualizations, Weights & Biases logging, and a Flask‑based graphical dashboard for easy pipeline configuration.
By Melika Shirian, Kianoosh Vadaei, Kian Majlessi, Audrina Ebrahimi, Peyman Adibi, Hossein Karshenas
The paper introduces beignet, a neural field architecture that replaces the random Fourier feature embedding in physics-informed neural networks with a trainable multi‑resolution Fourier feature pyramid. By using Fourier interpolation and spectral derivatives, beignet efficiently computes spatial derivatives and scales accuracy through the feature pyramid rather than the neural network size. Experiments show that beignet achieves more accurate PDE solutions with fewer parameters than existing PINN methods and can reach machine‑precision residuals on the inviscid Burgers blowup problem using the Adam optimizer.
By Brandon Zhao, Yixuan Wang, Jonathan T. Barron, Katherine L. Bouman, Dor Verbin, Pratul P. Srinivasan
HCPN-GCN is a graph-aware extension of Hierarchical Prototype Networks that replaces linear feature extractors with Graph Convolutional Networks and introduces cone-based prototypes with a diversity regularization objective. This design yields richer graph-aware representations, reduces prototype proliferation, and maintains near-zero forgetting. Experiments on six continual graph learning benchmarks show consistent accuracy improvements over the original HPN and other baselines while using roughly 30× fewer atomic prototypes.
By Sammuel R. Silva, Vander L. S. Freitas, Gladston Moreira, Eduardo J. S. Luz, Rodrigo Silva
The paper introduces Universal Attention, a transformer architecture that unifies and extends decay-based KV‑cache pruning mechanisms while preserving RoPE embeddings and Softmax attention. It achieves highly expressive, end‑to‑end trainable pruning that adaptively removes low‑contribution tokens, yielding up to 10× compression on standard datasets and 25× compression at 16k context length without sacrificing downstream performance. The approach outperforms existing state‑of‑the‑art baselines and unpruned models in both compression and generalization.
By Davis Wertheimer, Haochen Shen, Ahan Gupta, Derrick Liu, Yu Chin Fabian Lim, Mudhakar Srivatsa, Raghu K. Ganti, Minjia Zhang, Naigang Wang
The paper introduces TOPOGRAPHRAG-BENCH, a benchmark for evaluating multimodal Graph Retrieval-Augmented Generation (GraphRAG) systems on layout‑grounded evidence reasoning. It contains 2,024 questions derived from 201 long, visually rich documents, designed around three controlled topologies—single‑hop retrieval, bridge‑chain reasoning, and multi‑source synthesis—while employing counterfactual validation to ensure question integrity. Experiments show that multimodal GraphRAG outperforms text‑only and page‑level visual retrieval models, yet still struggles with incomplete visual‑textual alignment and multi‑unit composition, highlighting the need for GraphRAG systems that explicitly model document layouts and cross‑modal evidence alignment.
By Ruochi Li, Jianzhe Lin, Haoxuan Zhang, Haihua Chen, Junhua Ding, Edward Gehringer, Yang Zhang
The paper introduces PureVision, a geometry‑supervised framework for medical vision‑language models that tackles the challenge of interpreting multi‑phenotype lesions. It comprises PureEyes, which imposes geometric constraints to preserve anatomical hierarchies and phenotypic subclass relationships, and PureNeurons, which aggregates lesion‑specific anatomical and phenotypic evidence in a learned latent space. Experiments on LIDC‑IDRI, CBIS‑DDSM, and 3DReasonKnee show that PureVision enhances lesion grounding and phenotype characterization in visual question answering and radiology report generation.
By Hao Wang, Qiwei Zeng, Shuchang Ye, Jinghao Lin, Yuezhe Yang, Yige Peng, Jinman Kim, Lei Bi
EntroPrefill introduces a Renyi‑guided context pruning method for retrieval‑augmented generation, coupling a proposal mechanism with constraints on discarded attention mass. The approach includes a sink‑isolated, regularized head pooling strategy that balances specialization and worst‑head coverage, and derives a computable upper bound on token removal along with finite‑sample observer guarantees for adaptive pruning layers. The paper also establishes a conditional transformer perturbation bound with Lipschitz constants, a first‑token decision‑margin corollary, and a counterexample illustrating limits of shallow observations, while outlining a systems analysis that distinguishes query‑head unions, physical page allocation, and KV‑transfer payload to provide an arithmetic break‑even condition for pruning.
By Inbasekaran S
arXiv:2610. 09838v1 Announce Type: new Abstract: We present a framework for building and interpreting minimal transformer models.
By Raneem Mahajne, Toviah Moldwin
KGATE is a modular Python library for training knowledge graph embedding (KGE) models, built on PyTorch Geometric and TorchKGE. It allows users to assemble or plug in components such as initializers, encoders, decoders, losses, negative samplers, and evaluation metrics, and includes a preprocessing step to prevent data leakage. Benchmarks against six existing KGE libraries show that KGATE’s training time is comparable to the fastest while offering a broader set of features.
By Benjamin Loire, Galadriel Bri\`ere, C\'elia Brahimi, Antoine Toffano, Ana\"is Baudot
The paper introduces CARE, a method for certifying accelerated inference in vision‑language‑action models. CARE uses paired rollouts on a calibration set to guarantee that the risk of acceleration‑induced failure stays below a user‑specified budget, selecting the fastest certified accelerator or falling back to the reference policy. Experiments on LIBERO suites show that CARE achieves 9.0–10.8× speedups while preserving at least 85.8% of reference‑solved episodes with 95% confidence, and it outperforms non‑guaranteed selectors in both budget adherence and rollout efficiency.
By Rui Liu, Tong Zheng, Jindong Gu, Zhipeng Wang
arXiv:2610.09860v1 Announce Type: cross
Abstract: 4D point clouds acquired by permanent laser scanning (PLS) enable accurate high-frequency monitoring of surface change in dynamic topographic environ...
By Jiapan Wang, Daan Hulskemper, Mathilde Letard, Roderik Lindenbergh, Katharina Anders
The paper proposes that independently trained multimodal models often share a common geometric structure, allowing cross‑modal alignment without paired data. Using a Wasserstein Procrustes approach with a coarse geometric initialization, the authors align two disjoint embedding sets by estimating a single orthogonal map, achieving consistent alignment across datasets and modalities. While paired examples can further improve performance, the method already outperforms existing techniques in the few‑pair regime and enables text‑to‑image generation without paired data.
By Dominik Schnaus, Thomas Dag\`es, Daniel Cremers, Xi Wang, Phillip Isola
arXiv:2610.10135v1 Announce Type: cross
Abstract: Sparse attention mechanisms estimate attention over $n$ tokens using a small subset of keys. Many existing approaches use maximum inner product searc...
By Stepan Zharkov, Krish Singal, Ashwin Padaki, Alexandr Andoni
arXiv:2510.01335v2 Announce Type: replace
Abstract: The manifold hypothesis suggests that data lies on manifolds with smaller intrinsic dimension (ID) than their ambient dimension. However there is n...
By Aritra Das, Joseph T. Iosue, Victor V. Albert
The paper introduces a label‑free mixed‑precision post‑training quantization technique for text embedders called Quantize by Drift. It uses quantization‑induced representation drift—measured by re‑encoding a corpus after quantizing a module—to estimate module sensitivity without relevance labels. Across five development embedders, drift‑based ordering of mixed‑precision plans correlates strongly with held‑out retrieval quality (macro Spearman 0.911) and outperforms uniform precision under a hard byte budget, though it does not fully replace fine‑grained redistribution strategies.
By Hyojung Han, Jongmin Kim, Seung-Hun Jeon
The paper introduces DSReg, a method that provably recovers individual world latent variables without relying on reconstruction, decoders, or labels. It leverages a Structural Diversity condition—distinct dependency footprints of different latents—to achieve full identifiability up to signed permutation. DSReg can be applied post hoc to any linearly identified representation, preserving dense prediction while improving latent recovery across synthetic and real-world experiments.
By Yujia Zheng, David Klindt, Randall Balestriero, Bernhard Sch\"olkopf