GeoContext is a new vision‑language geolocation benchmark that introduces two tasks: GeoHint, where a model must localize an image given a coarse location hint, and GeoVerify, where a model must decide if an image was taken within 150 m of a claimed place. The benchmark builds a context ladder by stratifying nearby reference points by distance and referenceability, allowing the same image to be evaluated under varying context. Evaluation of five models on 109 sites in 30 cities shows that hint repetition is low, localization error grows with hint distance, and models struggle to achieve high discriminability in GeoVerify, with many false acceptances reported with high confidence.
By Yifan Zhang, Kai Wang
The paper introduces CLLPU, a multilingual benchmark for evaluating how well large language models can unlearn specific knowledge while controlling its propagation across languages. CLLPU defines two forgetting scenarios—common-goal forgetting, which requires suppression across all languages, and language-conditioned forgetting, which limits suppression to a single language. Using 800 knowledge-unit pairs and 72,000 QA instances in ten languages, the authors test six methods on Llama‑3.1‑8B‑Instruct and find that universal suppression often fails, while language‑specific suppression can unintentionally spread to other languages, highlighting the difficulty of propagation control in multilingual unlearning.
By Pengyang Shao, Chuanpeng Lu, Wei Qin, Yanzheng Jin, Xiaohao Liu, Xi Ai, Kenji Kawaguchi, Richang Hong
arXiv:2609.10008v1 Announce Type: new
Abstract: Composed video retrieval (CoVR) searches a gallery for the target video that realizes a natural-language modification of a source clip. However, at gal...
By Dmitry Demidov, Muhammad Zaigham Zaheer, Omkar Thawakar, Abdelrahman Mohamed Shaker, Rao Anwer
SCRIPTIOC-BENCH is a benchmark designed to evaluate how well large language models can statically extract indicators of compromise (IOCs) from script-based malware. It contains 634 manually verified JavaScript, PowerShell, and VBScript samples and covers four IOC types—URLs, domains, IP addresses, and filesystem artifacts—while distinguishing between directly exposed and encoded indicators. Experiments show that even the best models achieve only 65.4 F1, and a false‑positive taxonomy is introduced to analyze error patterns, with two mitigations (deterministic string utilities and task‑specific adaptation) improving precision and shifting errors toward sample‑grounded mismatches.
By Hanna Kim, Jian Cui, Minkyoo Song, Hwanjo Heo, Seungwon Shin, Kimin Lee, Xiaojing Liao
The paper presents new algorithms for fair k‑center clustering in Euclidean spaces, where a dataset is divided into groups and each group has a limit on the number of centers that can be chosen. A parameterized approximation algorithm achieves a 2.732 ratio, which is improved to 2.414 with exponential time in k. By integrating this into a one‑pass streaming framework, the authors obtain streaming approximations of 4.464 (improvable to 3.828) and a polynomial‑time streaming algorithm with a 4.732 ratio, further reduced to 4.42, surpassing previous state‑of‑the‑art results. Experiments confirm that these methods outperform existing approaches in clustering accuracy.
By Zeyu Lin, Chaoqi Jia, Longkun Guo, Chao Chen
The paper introduces Long-Short Term Advantage Estimator (LSTAE), a single‑stream reinforcement learning algorithm that replaces repeated within‑iteration trajectory sampling with historical experience for advantage estimation. LSTAE tracks each task anchor with a drift‑aware long‑term baseline and a short‑term state‑experience buffer, converting accumulated experience into multi‑granular credit signals while requiring only one rollout per anchor. Experiments on agentic and mathematical reasoning benchmarks show that LSTAE matches or surpasses strong group‑based baselines while significantly reducing rollout cost.
By Xinhao Yao, Lu Yu, Changhao Wang, Fengwei Teng, Yuyao Zhang, Qing Cui, Jun Zhou, Yong Liu
The paper introduces visual adaptations of counterfactual tests—vCT and vCCT—to evaluate whether chain-of-thought explanations in vision‑language models faithfully reflect the visual evidence driving predictions. Using these tests, the authors benchmark eight open‑source VLMs on two datasets and find that CoTs often fail to track visual evidence, sometimes omitting removed objects or mentioning them inconsistently. They also release two new datasets, Counter‑SNLI‑VE and Counter‑A‑OKVQA, consisting of image pairs that differ by a single object to facilitate further research.
By Bayar Menzat, Maximilian S\"uss, Ruizhi Wang, Benno Steinegger, Thomas Lukasiewicz, Oana-Maria Camburu
DPSF-Net is a dual‑prior spatial‑frequency network designed for real‑world remote sensing image dehazing. It combines hazy RGB images with dark channel prior maps as joint inputs, and incorporates a spatial‑frequency residual interaction block, a prior‑guided feature attention module, and a selective kernel complementary fusion module to reduce colour shift, structural distortion, and large‑scale haze. Experiments show that DPSF-Net achieves state‑of‑the‑art performance on the RRSHID benchmark while maintaining a favorable balance of restoration quality, parameter count, and computational complexity.
By Mei Lu, Shangliang Shao, Shanliang Yao
MV-STRIDE is a Multi‑View hierarchical Spatial Reasoning dataset that models dependencies among perception, scene understanding, and contextual reasoning to support 3D spatial cognition. It introduces a QA generation pipeline that enforces cross‑view constraints, producing multi‑level reasoning tasks with chain‑of‑thought supervision. Experiments show that training on MV‑STRIDE yields state‑of‑the‑art performance on multi‑view spatial benchmarks, enabling MLLMs to reason robustly across diverse viewpoints.
By Jin Xu, Xiaojian Huang, Zhuodong Luo, Zhihong Zhang, Xin Liu, Jiansheng Wei, Xinzhi Wang, Jie Zhao, Xuejin Chen
The paper introduces PLATOS, a scheduling strategy for Healthcare IoT in fog computing that balances power consumption and latency. PLATOS classifies tasks into priority, storage, and computational categories, then selects fog resources to minimize delay and energy use before allocating high‑performance nodes to high‑priority tasks. Simulations show an 18.72% reduction in energy use and an 8.65% decrease in latency compared to existing methods.
By Mohammed Alaa Ala'anzy, Zulfiqar Ahmad, Zhanar Mukash
The paper investigates adding Greek language support to a Cosmos3 vision‑language‑action robot policy using only machine‑rephrased instructions and no architectural changes. It finds that many evaluation metrics can give misleading results, and that multilingual training with Greek yields a modest 6.7‑7.1 point advantage over a control, reaching about 40% of English performance. The study also shows that overfitting to a single translator’s phrasing can be mitigated by training on multiple phrasings, while warm‑starting from a language‑adapted world model or unfreezing the text tower actually harms performance.
By Ayoub Kirouane, Georgios Giaples, Christos Petrocheilos
The paper introduces Mobility Stream-Structure Synergy (MoSS), a method that fuses two complementary views of mobility data—an hourly inflow/outflow Sequence view and a Structure view derived from zigzag persistence diagrams—to capture temporal dynamics and evolving regional connectivity. MoSS employs a synergy module that extracts higher‑order representations from the co‑occurrence of these views, moving beyond additive fusion. Experiments on New York City and Chicago demonstrate that MoSS outperforms existing baselines on three downstream tasks using only mobility data.
By Namwoo Kim, Jeeyun Chang, Kanghoon Lee, Yoonjin Yoon
The paper introduces Feedback‑Enriched Environments (FEEs) as a new approach to training large language models as autonomous agents for long‑horizon tasks. By shifting from action guidance to observation enrichment during later stages of exploration, FEEs improve performance across SciWorld and BFCL benchmarks with various Qwen3 model scales and RL algorithms. The study shows that FEEs stabilize training, promote proactive exploration, embed environmental guidance into policy weights, and highlight intra‑group feedback consistency as key for stable optimization.
By Hongbang Yuan, Zhuoran Jin, Yixin Cao
The paper introduces BIFTA, a Brain‑Inspired Few‑Shot Tactile Adaptation framework that enables a frozen encoder to adapt quickly to an unknown tactile sensor using only a small labeled support set. It preserves pretrained representations via dual‑view statistical memory, builds support‑conditioned spectral graphs to correct sensor‑dependent feature neighborhoods, and employs uncertainty‑gated recurrent propagation to reinforce reliable cross‑query evidence. Benchmarks on three tactile datasets demonstrate that BIFTA dramatically improves adaptation performance, achieving an 87.09% mean Sparsh accuracy on SITR with just 10% labeled data—an increase of 47.22 percentage points over the best prior method.
By Boheng Liu, Ziyu Li, Xia Wu
CausalChapter is a framework designed to improve chaptering of long instructional videos by using interventional dependency modeling. It introduces a Local Dependency Shift module that identifies boundary drops between adjacent temporal windows and a Cross-Segment Support Selection module that re‑ranks historical contexts to enhance chapter description generation. Experiments on long‑video chaptering benchmarks demonstrate that CausalChapter improves boundary localization, chapter description quality, and cross‑chapter coherence.
By Xinran Duan, Guozhang Li, Yaoyao Zhong, Mei Wang, Lizhi Wang, Hua Huang
arXiv:2609.06289v1 Announce Type: cross
Abstract: As large language models (LLMs) are increasingly deployed in alignment-sensitive contexts, activation steering has emerged as a lightweight, inferenc...
By Mohammad Mahdi Abootorabi, Armin Saghafian, Ali Bazshoushtari, Hamid Rezaei, EunJeong Hwang, Vered Shwartz, Parvin Mousavi, Purang Abolmaesumi
arXiv:2609.06394v1 Announce Type: cross
Abstract: Massive datasets in modern machine learning have made data reduction a central challenge, particularly for clustering tasks where memory and computat...
By Diptarka Chakraborty, Satyaki Mukherjee, Gaurav Vallabhdas Revankar, Hoang-Son Tran
arXiv:2609.07312v1 Announce Type: new
Abstract: This paper proposes a robust decentralized personalized federated learning method R-DPFL, that enables clients to reduce the impact of Byzantine attack...
By Xiao Ma, Hong Shen, Hui Tian, Wenqi Lyu, Wei Ke
arXiv:2609.05806v1 Announce Type: new
Abstract: Emotion Recognition in Conversations (ERC) aims to identify speakers' emotions in multi-turn dialogue. Accurate emotion recognition can support a wide...
By Amir Ben Khalifa, Fanny Bezancon, Amine Trabelsi, Bessam Abdulrazak
arXiv:2609.05824v1 Announce Type: new
Abstract: Large language model (LLM) agents increasingly rely on external skills, but routing user requests over large skill registries is difficult because many...
By Wang Wei, Tiankai Yang, Samyadeep Basu, Hongjie Chen, Yue Zhao, Zhengzhong Tu, Xiyang Hu, Franck Dernoncourt, Ryan A. Rossi, Hoda Eldardiry