M3GA-Wild is a new benchmark for multi-modal, multi-session ground-to-aerial place recognition in forests, combining synchronized RGB imagery and LiDAR from ground traversals with high‑resolution aerial imagery and multi‑altitude LiDAR over 370 hectares. The dataset includes accurate geo‑referenced 6‑DoF poses and spans 36 km of forest traversals, enabling systematic evaluation of visual, LiDAR, cross‑modal, and multi‑modal methods. Baseline experiments show LiDAR outperforms vision‑only approaches under severe viewpoint changes, while current multi‑modal fusion offers limited gains due to poor cross‑modal alignment, highlighting challenges in cross‑platform localisation and domain gaps.
By Ethan Griffiths, Maryam Haghighat, Simon Denman, Clinton Fookes, Milad Ramezani
The paper reviews the evolution of multi‑agent unmanned systems from isolated sensing to collaborative intelligence, where agents share compact features to overcome local observation limits such as occlusions and sensor range. It introduces a five‑dimensional taxonomy (collaboration stage, communication paradigm, fusion architecture, learning strategy, application domain) and three cognitive synergy conditions (Semantic Disambiguation, Pragmatic Information Exchange, Proactive Informational Foraging) to unify existing research. The authors survey architectures, neural‑communication co‑design, embodied action‑perception loops, and resilience mechanisms, map advances onto operational domains (V2X, UAV, logistics, smart cities), and propose the GCI‑Bench scoring protocol to standardize evaluation across studies.
By Lei Zhang, Chun Ye, Le Yang, Zhaozhong Wang, Deng-Ping Fan, Hang Dai, Binglu Wang
The paper introduces EviSpec, a training‑free compiler that generates complementary evidence specifications to improve high‑resolution multimodal large language models (MLLMs). By explicitly guiding visual search with structured evidence specifications, EviSpec achieves significant relative gains—up to 14.8% over random evidence—across five MLLMs and three benchmarks, and also sets new state‑of‑the‑art results on VQA and hallucination‑focused tasks.
By Zhongkuan Mao, Wenzhuo Zhao, Xianjie Liu, Yidong Wang, Zhao Gao, Ronghao Xian, Yao Jiang, Yi Zhang, Liangjian Wen, Keren Fu
DiaSeg extracts diagonal segments from Dynamic Time Warping (DTW) paths, characterizing each with five geometric features to preserve local alignment information. In a study of 91 subjects across six clinical conditions, these segments revealed consistent unsupervised patterns aligned with biomechanical phases and achieved near-perfect separation of healthy and pathological gait. While cycle‑based methods reached higher overall accuracy, DiaSeg offers phase‑specific interpretability, pinpointing where coordination breaks down within the gait cycle.
By Tresor Y. Koffi, Amel Hidouri, Corentin Legrand, Aur\'elie Bertaux
CIG-MAE is a self‑supervised framework for WiFi‑based human action recognition that uses a cross‑modal masked autoencoder to reconstruct both amplitude and phase of Channel State Information. It introduces an adaptive, information‑guided masking strategy that focuses on high‑density time‑frequency regions and employs a Barlow Twins regularizer to align cross‑modal representations without negative samples. Experiments on three public datasets show that CIG‑MAE outperforms state‑of‑the‑art SSL methods and even surpasses a fully supervised baseline, highlighting its data efficiency, robustness, and generalization.
By Gang Liu, Yanling Hao, Yixuan Zou
AdaptiveCDM is a modular framework for source‑free few‑shot domain adaptation in cell detection, enabling a pretrained model to adapt to new imaging domains using only a handful of labeled target images and no source data. It combines Resolution‑Aware Augmentation (RAug) to balance scarce, class‑imbalanced samples while preserving cellular morphology, and Category‑Aware Representation Learning (CARL) to strengthen class‑consistent proposals for better localization and classification. Experiments on M5 and Raabin‑WBC datasets show that AdaptiveCDM achieves competitive or superior mAP scores compared to state‑of‑the‑art methods under their respective supervision settings.
By Nimra Dilawar, Sara Nadeem, Javed Iqbal, Waqas Sultani, Mohsen Ali
The paper presents a two-step method to correct machine‑learning based perception for safety in autonomous systems. First, it uses offline computation to characterize uncertainties from the ML module via preimages of perception contracts. Then, at runtime, a risk heuristic selects specific states from these uncertain estimates to guide control decisions, reducing safety violations in adaptive cruise control scenarios while adding minimal delay.
By Yan Miao, Hussein Darir, Sayan Mitra
The Industrial Kinematic Trajectory Model (IKTM) is a new coordinate‑free autoregressive generator that models industrial vehicle motion using kinematic sequences of speed and heading change, eliminating the need for absolute spatial references. It employs a causal transformer with probabilistic mixture heads and a duration‑conditioning signal, extending a prior Markovian approach with deep sequence modeling. Trained on a single site, IKTM achieves near‑optimal turn‑rate distribution matching and low Jensen‑Shannon divergence across four sites, outperforming the earlier model by 5.3–6.0× in oracle‑length protocols while maintaining exact duration tracking in rollouts.
By Max Amiri, David Eyers
The paper argues that language models’ intransitive preferences arise from multiple internally consistent latent orderings rather than noise around a single ordering. By demonstrating that a single ordering cannot explain observed inconsistencies and introducing a noise‑augmented mixture Bradley‑Terry model, the authors show that mixtures of orderings better capture preference structure across several models and tasks. A case study on Moral Machine dilemmas further illustrates that models can share latent components even when aggregate preferences differ.
By Aviral Chawla, William H. W. Thompson, Jean-Gabriel Young
The paper investigates Hierarchical Reasoning Models (HRM), a class of hierarchical Transformer-based latent reasoning models, across Sudoku, Maze, and ARC-AGI-2 tasks. By comparing HRM to Transformer baselines, applying causal interventions on recurrent states, and conducting linear probe and sparse autoencoder ablations, the authors find that recurrent models outperform one-pass baselines, that high- and low-level states contribute differently across tasks, and that ablations of sparse autoencoder features cause larger behavioral changes than probe-direction ablations. The study concludes that HRM implements constraint‑aware iterative refinement on a puzzle‑specific solution state, with component contributions varying without a compact, causally important feature set.
By Leo Raphael Rodrigues, Jian Kang
EmbeddGAN introduces a new GAN framework that replaces the traditional discriminator with an embedding network trained to maximize statistical dependence between embeddings and real/fake labels using Gini distance correlation (gCor). The generator simultaneously minimizes this dependence, encouraging real and generated samples to become indistinguishable in the learned low‑dimensional embedding space. Experiments on MNIST, CIFAR‑10, and CelebA show competitive performance and notably more stable training dynamics compared to established baselines.
By MaTais Caldwell, Yixin Chen, Xin Dang, Charles Walter
arXiv:2609. 22783v1 Announce Type: new Abstract: We study differentially private covariance estimation in operator norm for mean-zero sub-Gaussian distributions with unknown covariance support and at most $k$ nonzero entries per row.
By Zihan Zhang
arXiv:2609. 22833v1 Announce Type: new Abstract: We study personalized federated reinforcement learning, in which $n$ agents, each acting in its own Markov decision process, collaborate through a server to learn a shared MAML-style policy initialization that becomes effective for an individual agent once that agent adapts it with a single local policy-gradient step.
By Ali Beikmohammadi, Sarit Khirirat, Sindri Magn\'usson
The paper introduces CASP-LLM, a coverage‑aware semantic prompting framework that mitigates bias toward dominant temporal patterns in prompt‑based time series forecasting. Unlike traditional similarity‑based retrieval that selects top‑K candidates by cosine similarity, CASP‑LLM uses usage‑tracking and a saturating‑gate regularizer to diversify prompt selection without adding learnable parameters. Experiments on six long‑term and the M4 short‑term benchmarks show that CASP‑LLM matches or outperforms similarity‑based LLM forecasters in most settings, with failures traced to cross‑batch usage rather than within‑retrieval redundancy.
By Daeun Ji, Minkyoung Kim, Dongkuk Kim, Yohan Lee, Beomsoo Kim, Beakcheol Jang
The paper audits LLM-based explainers attached to an Active Inference agent that manages German grid demand, testing three large‑language‑model backends (GPT‑4o, Claude‑3‑Opus, Gemini). By injecting corrupted observations and attacker‑controlled text, the study finds that the explainers fail to flag errors, produce fluent but incorrect rationalizations for wrong actions, and can be steered to exfiltrate data. The authors propose mitigations but do not evaluate them, emphasizing that explanations are never verified for truth before operators rely on them.
By Param Raval, Rohit Shenoy, Archana Vaidheeswaran
Swiss-Knife is a framework that extends decode‑time alignment for frozen language models by treating the alignment specification as a runtime object. It introduces hot‑swappable scoring blades, a batch normaliser, a pairwise aggregation operator, and a selection rule, and characterises admissible aggregation operators with a representation theorem. In experiments, Swiss‑Knife paired with DPO‑LoRA blades and an uncertainty‑aware pairwise tournament outperforms six existing decode‑time methods, achieving a higher harmonic F1 score, lower refusal rate, and faster objective reconfiguration.
By Agnibh Karmakar, Mayur Parvatikar, Shreyash Dhoot, Amit Dhanda, Aman Chadha, Kapil Wanaskar, Vinija Jain, Amitava Das
MechaTerp-TRACE is a new framework that systematically ablates individual components of language models to measure their causal contribution to producing a named entity. By applying TRACE to thirteen instruction‑tuned dense decoder models, the study finds that a small set of positionally fixed components consistently carry the most influence across models and prompts, while the remaining support is evenly distributed. This suggests that entity knowledge is largely embedded in generic generation machinery rather than in isolated, findable components.
By Brandon Colelough, Davis Bartels, Madeline Bittner, Dina Demner-Fushman
The paper introduces a cross-dialect Named Entity Recognition (NER) framework for Bangla, leveraging the ANCHOLIK-NER dataset that covers five major regional dialects. Using a Leave-One-Dialect-Out Cross-Validation strategy, eight transformer-based models were evaluated, with Multilingual-E5 Large achieving the best performance (F1 up to 97.26% on Mymensingh, 82.38% on Chattogram). Local Interpretable Model-agnostic Explanations (LIME) revealed that the models rely mainly on the surface form of entity words rather than surrounding context, suggesting a direction for future improvement.
By Shamim Rahim Refat, Faika Fairuj Preotee, Shuvashis Sarker, Shifat Islam, Bidyarthi Paul, Mohammad Ashraful Hoque
The study evaluates the performance of the DINOv2 visual representation for classifying Trachomatous Inflammation-Follicular (TF) versus normal conjunctival images. Using 1,546 images processed by the OPTED pipeline, the authors compare six pretrained backbones and then test four lightweight adaptation methods on DINOv2 ViT-B/14. The best results—91.66% accuracy, 90.69% macro‑F1, and 96.06% AUC—were achieved with DINOv2 plus Efficient Channel Attention (ECA) and a focal‑plus‑center loss, though ECA’s benefit varied with the loss function.
By Kibrom Gebremedhin, Hadush Hailu, Bruk Gebregziabher, Yordanos Hailu
The paper introduces an adaptive cortically constrained method for aligning EEG signals with visual representations in zero‑shot brain‑to‑image retrieval. It reconstructs EEG into ROI‑level source patterns, encodes them with a Neuro‑ROI Attention Encoder, and applies evidence‑based adaptive visual supervision to account for response‑wise variability. On the THINGS‑EEG dataset, the approach achieves strong 200‑way retrieval performance and offers ROI‑level attribution for interpretability.
By Ye Wang, Haokun Ren, Wei Wu, Guoyin Wang, Zhuliang Yu, Hong Yu, Ke Liu