The paper introduces an unsupervised peer-relative representation learning framework to detect energy inefficiencies in mobile network sites. It extends Minimum Distortion Embedding with an energy-based repulsion term, creating a low-dimensional space that preserves structural similarity while highlighting sites with anomalously high energy use. The resulting anomaly scores help operators prioritize field investigations and achieve large-scale energy-efficiency optimization.
By Eliud Nyakweba Koto, Jaco du Toit, Adham Stoltz, Johan du Preez
arXiv:2609. 03487v1 Announce Type: cross Abstract: Knowledge graph embedding (KGE) demonstrates its effectiveness for predicting missing links in knowledge graphs (KGs) by projecting entities and relations into a low-dimensional vector space.
By Junsik Kim, Kangil Kim
The paper introduces CS‑VAR, a Cross‑Session Evidence‑Aware Retrieval‑Augmented Detector that uses a lightweight, domain‑specific model for fast session‑level risk inference in live streaming. The model is trained with guidance from a Large Language Model that reasons over retrieved cross‑session behavioral evidence, enabling the small model to recognize recurring risk patterns and perform structured risk assessment. Experiments on large industrial datasets and online validation show that CS‑VAR achieves state‑of‑the‑art performance while providing interpretable, localized signals for real‑world moderation.
By Yiran Qiao, Xiang Ao, Jing Chen, Yang Liu, Qiwei Zhong, Qing He
The paper introduces PACE, a dataset that tests whether personalized assistants can detect hidden conflicts in user requests by integrating implicit, egocentric knowledge from a knowledge base. PACE pairs user requests with persona-specific facts, requiring models to retrieve contextual evidence before deciding if a request is inappropriate. The authors also propose PaceMaker, a multi-agent system that improves conflict detection by coordinating query reformulation, multi-hop graph traversal, and conflict-aware filtering, outperforming existing methods on the PACE benchmark.
By Yoojin Kim, Jihyoung Jang, Hyounghun Kim
EmbedTalk introduces per‑Gaussian embeddings to drive speech‑driven facial deformations in real‑time talking head synthesis, replacing traditional tri‑plane encodings. This approach improves rendering quality, lip synchronisation, and motion consistency compared to prior 3D Gaussian Splatting methods while producing more compact models that run at 60+ FPS on a laptop GPU. The technique demonstrates competitive performance against state‑of‑the‑art generative models.
By Arpita Saggar, Jonathan C. Darling, Duygu Sarikaya, David C. Hogg
R$^{2}$Adapter is a lightweight plug‑in that dynamically routes user queries between vanilla and graph‑based Retrieval‑Augmented Generation (RAG) systems. By sending only those queries that truly benefit from graph reasoning, it cuts graph‑retrieval overhead by up to 59% while keeping answer accuracy comparable. The adapter also rewrites uncertain graph‑routed queries to better expose multi‑hop reasoning needs, improving retrieval quality without extra supervision.
By Yucan Guo, Miao Su, Saiping Guan, Long Bai, Zhongni Hou, Zixuan Li, Xiaolong Jin, Jiafeng Guo, Xueqi Cheng
The paper introduces CRAW, a codec‑robust audio watermarking framework designed to embed imperceptible signals into synthetic speech. CRAW enhances robustness against neural re‑synthesis, codecs, denoisers, and vocoders while preserving high perceptual quality through distortion‑aware training, attention‑based pooling, perceptual masking, and error‑correcting codes. Experiments show CRAW outperforms existing post‑hoc watermarking methods in robustness without compromising audio quality.
By David Chernin, Ethan Fetaya
Reflect‑SQL is a new framework for converting natural language into SQL queries. It tackles challenges such as large, obscure database schemas, poor table and column retrieval, and syntactically or logically flawed SQL by using a multi‑stage self‑reflection approach. The system iteratively refines queries and SQL through feedback loops driven by an LLM‑as‑a‑judge, achieving 72.03% execution accuracy on the BIRD benchmark, outperforming existing baselines.
By Anupreksha Jain, Manish Shrivastava
The study maps tree species across Denmark using National Forest Inventory plots and Earth Observation data, comparing manually engineered spectral‑temporal features (STF) from Sentinel‑1 and Sentinel‑2 with embeddings from the foundation models TESSERA and AlphaEarth. Random forest, XGBoost, and MLP classifiers were evaluated, with the STF‑based MLP achieving the highest macro F1 scores for pure and mixed stands. The best model was applied nationally to produce a 10 m resolution tree species map, achieving 79.9% area‑adjusted accuracy and released as an open‑access product.
By Alkiviadis Koukos, Spyros Kondylatos, Thomas Nord-Larsen, Lotte Nyborg, Christian T{\o}ttrup, Kenneth Grogan
FinRAG-QA is a new benchmark dataset for financial question answering, featuring 999 practitioner-curated questions on 10 standardised indicators drawn from 209 annual and Pillar 3 reports of 24 major European and U.S. banks between 2019 and 2023. The dataset focuses on cross‑institutional retrieval over documents averaging 198k words, making it longer than any existing financial QA resource. Experiments on a multi‑stage Retrieval‑Augmented Generation pipeline show that contextual chunk enrichment and a retrieval‑optimised embedding model significantly improve NDCG@10, while a reasoning‑optimised generator boosts answer accuracy from 44.6% to 79.0% when the correct document is retrieved.
By Arianna Miola, Bruno Spaccavento, Lorenzo Silotto, Marco Bianchetti, Luca Cagliero
The Dice Roll Method is a standardized protocol for auditing large language model brand recommendations through repeated queries. It decomposes total response variance into sampling, prompt‑phrasing, run‑to‑run, and model‑version components, and uses a negative‑binomial mixed model, Cliff’s delta, and bootstrap techniques to guide iteration counts. The study identifies three iteration tiers—exploratory (n=5), confirmatory (n=10), and rigorous (n=15)—and recommends a compact battery of four complementary metrics for robust evaluation.
By Dmitrij \.Zatuchin
The paper introduces a physics‑informed graph attention network that directly operates on the tetrahedral mesh used in TCAD simulations of FinFET devices. By predicting electrostatic potential and quasi‑Fermi levels at every mesh node and training with both data loss and finite‑volume current‑continuity residuals, the surrogate retains the underlying carrier‑transport physics while achieving size generalization. Benchmarks against Sentaurus Device show sub‑volt RMSE for the drift‑diffusion fields and a per‑design throughput that is orders of magnitude faster, enabling rapid Pareto‑front exploration of large multi‑fin arrays that would otherwise be prohibitively slow to simulate.
By Leonid Popryho, Ayoub Sadeghi, Inna Partin-Vaisband
The paper introduces TRIM, a black‑box defense for backdoor attacks in computer vision models. TRIM identifies and removes malicious trigger regions at inference time using region‑based segmentation, adaptive trigger discovery via inpainting and diffusion, and selective purification, without needing model internals, training data, or clean samples. Experiments on various datasets and trigger types show TRIM reduces attack success rates to as low as 1.16% while maintaining high clean accuracy.
By Ahmed Abdelnaby, Mohamed Elmahallawy
The study evaluates self‑supervised learning (SSL) models pretrained on ImageNet‑1k and the Human Protein Atlas (HPA) Field‑of‑View (FOV) for protein localization in microscopy images. DINO‑based Vision Transformer backbones pretrained on either dataset transfer well to the OpenCell dataset, achieving strong performance even without fine‑tuning and improving further when fine‑tuned (0.704 ± 0.027 macro F1 on 17 classes). At the single‑cell level, the HPA‑pretrained model outperforms others in k‑nearest‑neighbor classification across all neighborhood sizes (macro F1 ≥ 0.515).
By Ben Isselmann, Dilara G\"oksu, Heinz Neumann, Andreas Weinmann
The paper introduces TEmBed, a unified benchmark for evaluating tabular embeddings across four representation levels—cell, row, column, and table—using a diverse set of models. It demonstrates that the best model depends on the specific task and representation level, providing practical guidance for selecting embeddings in real-world applications. The study aims to facilitate the development of more general-purpose tabular representation models.
By Liane Vogel, Kavitha Srinivas, Niharika D'Souza, Sola Shirai, Oktie Hassanzadeh, Horst Samulowitz
The paper investigates using synthetic natural-language descriptions to contrastively pretrain small transformer encoders for code representation. By pairing generated descriptions with code in a dual-encoder setup during training and discarding them at inference, the authors achieve significant improvements over traditional pretraining baselines on most evaluated tasks. When fine‑tuned, these models match or surpass much larger zero‑shot models and remain competitive with execution‑aware supervision, indicating a scalable alternative for code embeddings.
By Kenneth Paulsen, Florian Tambon, Mike Papadakis, Shin Yoo
The paper introduces a prompt‑engineering framework that personalizes large language model (LLM) teaching assistants across disciplines by tailoring responses to six learner‑specific dimensions, creating 96 distinct learner profiles. It also analyzes student queries through Bloom’s Taxonomy to gauge cognitive complexity, encoding both learner attributes and cognitive assessments into structured prompts that condition the LLM without retraining. Experiments using NLP metrics and a small human study demonstrate that this approach yields perceptible differences in response style and structure, with statistical evidence linking specific learner attributes to measurable changes.
By Saptarshi Basu, Sandeep Kakar, Ashok Goel
arXiv:2609. 04046v1 Announce Type: cross Abstract: What can a single layer of self-attention compute?
By Rajmohan Rajaraman, Ravi Sundaram, Amanuel Tesfaye
Spruce is a system that enables scalable private outsourced retrieval by learning compact binary embeddings and using efficient Hamming-distance computation under a two‑server multi‑party computation protocol. It replaces costly corpus‑wide embedding scoring with a fixed‑radius protocol that avoids multi‑round candidate selection, and introduces private cluster pruning and a one‑core dealer to reduce computation and eliminate OT preprocessing bottlenecks. Across corpora of 383K–5.42M documents, Spruce maintains original search quality while achieving up to 31.5× higher throughput and reducing query times to a few seconds.
By Peichun Hua, Yunming Xiao
LexIssue introduces a benchmark for identifying disputed legal issues in Chinese civil litigation, comprising 430 real‑world cases and 1,303 expert‑annotated issues. The dataset is built around a hierarchical schema that links free‑form issue descriptions to structured legal categories, enabling two complementary tasks: issue generation and issue classification. A retrieval‑augmented knowledge base covering 27 causes of action and 441 issue entries is provided, and experiments show that incorporating this knowledge consistently improves model performance on the tasks.
By Huiyuan Xie, Yuqin Huang, Zhicheng Hao, Yida Cai, Shaochun Wang, Zhenghao Liu, Yuxiao Ye