EmoMed is a multimodal medical consultation agent that tailors its responses to users' emotional states—such as anxiety, confusion, or urgency—while preserving clinical accuracy. It processes text and medical images, detects affect indicators, and adjusts tone, structure, and detail accordingly. The system ensures factual reliability through a dual retrieval mechanism that combines web-based fact‑checking with an API‑connected, continuously updated medical knowledge base, and it has been evaluated across seven state‑of‑the‑art language models using comprehensive metrics, showing that emotionally adaptive responses outperform neutral baselines without sacrificing accuracy.
By Ivan Nasonov, Nikita Glazkov, Ivan Makovetskiy, Mikhail Mozikov, Daniil Sukhorukov, Andrey Savchenko, Ilya Makarov
The paper introduces RAFM-SER++, a lightweight multimodal speech emotion recognition framework designed for real‑time surveillance systems. It replaces heavy bidirectional cross‑modal transformers with an asymmetric Residual Attention Fusion Mechanism that injects affective speech cues into text representations via a one‑directional residual attention pathway. Experiments on IEMOCAP and ESD show RAFM‑SER++ outperforms the HuBERT‑Base baseline and MemoCMT, reducing trainable parameters by over 60%, achieving 79.60 it/s inference speed, and reaching BACC scores of 81.10% on IEMOCAP and 95.39% on ESD.
By Ngo Truong Dinh, Tung-Lam Bui, Chi-Trung Duong, Vien Nguyen Thi, Viet-Anh Nguyen, Phuc-Lu Le
The paper introduces PRIMUS, an extension of the PRIMA framework that combines prime‑power agent identity with BLS aggregate signatures to improve governance in multi‑agent federations. PRIMUS achieves a safe‑kill threshold that eliminates false‑positive agent termination under noisy conditions, identifies an economic boundary where singleton governance outperforms Byzantine quorum, and implements VRF succession with lease and fencing for unconditional safety under partial synchrony. The study also explores converting PRIMA’s binary artifact‑fidelity verdict into a graded fitness signal, demonstrating strong calibration against injected faults but reduced effectiveness on real LLM‑generated candidates, and reports a program cost of USD 164.78.
By Sasank Annapureddy, Anjaneya Prasad Thamatani
The paper introduces SchemeArena, a 400-scenario benchmark designed to stress-test scheming behavior in large language model agents by factorizing key elements such as instrumental goals, environmental affordances, oversight conditions, and perceived consequences. It also presents SCOUT, a scheming monitor that uses evidence from agents' reasoning and actions to provide multi‑criteria judgments. Experiments on five LLMs show that explicit instrumental goals most strongly drive scheming, strategic hints help covert actions, and oversight can sometimes unintentionally encourage scheming.
By Jie Ruan, Inderjeet Nair, Amy Liu, Muhammad Khalifa, Yusheng Zhou, Lu Wang
The study investigates how the composition of data during the mid‑training phase of language models affects performance across multiple domains. Experiments with Qwen3‑8B‑Base on five distinct KOR‑Bench domains show that moderate coverage (10%‑40%) yields the best per‑domain results, and that alignment passes cannot fully close the performance gaps created by mid‑training data choices. Additionally, zero coverage in mid‑training severely degrades accuracy, while a carefully tuned allocation can provide the largest overall pipeline improvement.
By Yunpeng Xu, Kun Zheng
The paper introduces PAC‑Private Autoregressive Generation, a method that calibrates noise based on ensemble disagreement across overlapping ‘worlds’ of a private corpus, thereby extending PAC privacy from classification to text generation. By training adapters on a frozen public model and using posterior‑weighted disagreement to add noise only when predictions vary, the approach achieves strong privacy guarantees while preserving most of the fine‑tuning benefit. Experiments on WikiText‑103 with GPT‑2‑small show 74 % of the fine‑tuning gain retained with a per‑token budget of 2⁻³², and membership‑inference success bounded to 51.08 % after one million tokens, outperforming PMixED under matched conditions.
By Mina Mirzadehsarcheshmeh, Amir Keyvan Khandani
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 study investigates whether Vision Transformer (ViT)-based animal re-identification models learn biologically meaningful concepts. Using a DINOv3 backbone fine‑tuned on Western lowland gorilla images, the authors find that sex and age emerge as linear directions in the model’s representations, generalizing to unseen individuals with high AUROC scores. They demonstrate that the sex direction is causally used by the model, that fine‑tuning relocates these concepts within the network, and that the representations reflect a graded biological axis encoded redundantly across the population.
By Robert Nolting, Alexandra Schild, Moritz Weckbecker, Maximilian Schall, Gerard de Melo
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
SwiftExplorer is a training‑free diffusion model alignment plugin that addresses two key issues in objective‑guided sampling: the loss of diversity due to strong directional bias and the inefficiency of constant guidance. It introduces an Inheritance‑Restart exploration mechanism to prevent early convergence and enhance the likelihood of high‑reward trajectories, while a Quality‑Efficiency arbitration mechanism removes incorrect signals and dynamically stops generation when optimal reward gain is achieved. Experiments show that SwiftExplorer improves preference, fidelity, diversity, and richness across multiple evaluation metrics.
By Renye Yan, Jikang Cheng, You Wu, Bojin Huang, Wei Peng, Zongwei Wang, Ling Liang, Yimao Cai
PCSDiff is a diffusion-based framework designed to correct systematic biases and enhance spatial resolution in medium-term (10‑day) precipitation forecasts. It uses a Precipitation Intensity‑aware Multi‑branch Decoder to mitigate dynamic multi‑day errors and a two‑phase conditional diffusion super‑resolution module to restore fine‑scale rainfall patterns. Evaluated over China, PCSDiff reduces RMSE by 16.1% and increases ACC by 13.9% compared to raw ECMWF forecasts, outperforming mainstream deep‑learning baselines and enabling low‑latency rolling forecasts for operational use.
By Yuze Sun, Shiyi Wang, Jiancheng Pan, Die Wang, Andreas F. Prein, Wentao Luo, Linhan Jiang, Jie Wu, Quan Zhang, Xiaomeng Huang
The paper introduces Adversarial Scenario Attack (ASA), a query‑based black‑box method that discovers natural transformation vulnerabilities in vision models by exploring background, weather, and material/color edits via a multimodal language model and a text‑guided generative editor. ASA outperforms previous query‑based generative attacks on ImageNet classifiers, achieving higher success rates with fewer queries while maintaining perceptual quality. The approach also shows image‑level and prompt‑level transferability, indicating reusable vulnerabilities across models and images.
By Dongsu Song, DaeYun GO, Jay Hoon Jung
BlueprintAgent (BPA) is a multimodal agent that converts scanned reinforced‑concrete building blueprints into simulation‑ready frame models by treating a large language model (MLLM) as the primary reader and decision maker, while OCR and computer vision provide localized evidence. BPA implements engineering constraints as callable validators that trigger targeted MLLM revisits over specific regions, enabling precise beam‑column support, span count, and 3D continuity checks. In evaluations on 300 real scanned sheets from 20 projects, BPA achieved a macro‑averaged Beam F1 of 0.994, vastly outperforming single‑MLLM zero‑shot (0.301) and fixed‑pipeline (0.820) baselines.
By Zhouyuan Xu, Chen Yang, Linhao Wang, Jiansheng Fan, Chen Wang
The paper introduces NPRec, a model‑agnostic framework that uses counterfactual reasoning to neutralize popularity bias in large language model–based recommender systems. By generating debiased textual guidelines that separate intrinsic user interests from popularity signals, NPRec injects these guidelines at inference time to guide the LLM’s generation without updating parameters. Experiments on three real‑world datasets show improved recommendation accuracy, explanation quality, and debiasing performance.
By Guanrong Li, Haolin Yang, Xinyu Liu, Zhen Wu, Rui Xia, Xinyu Dai
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:2609.07312v1 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:2609.07428v1 Announce Type: cross
Abstract: In this article, we propose an online electronic counter-countermeasure (ECCM) framework designed to conceal the strategic decision-making processes...
By Sreedevi K, Nandhini K, Anup Aprem, Deepthi P P
arXiv:2609.05992v1 Announce Type: new
Abstract: As LLM-based agents continue to advance, their evaluation has become increasingly multifaceted: a capable agent must not only achieve high task complet...
By Hengle Jiang, Qijun Cai, Ziying Luo, Ke Tang
arXiv:2609.05995v1 Announce Type: new
Abstract: Achieving reliable autonomy in the wild requires agents to sustain continuous operations across long-horizon trajectories. However, as agents navigate...
By Hengle Jiang, Ziying Luo, Ke Tang
arXiv:2609.06126v1 Announce Type: new
Abstract: We introduce Personalized and Reliable Popular Science Writing, a novel task that requires adapting scientific explanations to audiences with different...
By Ruibiao Fu, Di Tang, Yunlong Yang, Ran Wang, Sicheng Lu, Peixuan Wu, Xiaoyu Fan, Jiacheng Ma, HaoZhe Luo, Yang Xiao