The paper introduces T2M Mamba, a text‑to‑motion generation model that addresses two key shortcomings of prior work: the independent treatment of motion periodicity and keyframe saliency, and sensitivity to paraphrased text. It achieves this by employing a Periodicity‑Saliency Aware Mamba module that estimates keyframe weights with Density Peaks Clustering and motion periodicity via FFT‑accelerated autocorrelation, and a Periodic Differential Cross‑modal Alignment Module (PDCAM) that improves alignment between textual and motion embeddings. Experiments on HumanML3D and KIT‑ML datasets demonstrate significant performance gains, achieving an FID of 0.068 and improvements across all other metrics.
By Xingzu Zhan, Chen Xie, Honghang Chen, Yixun Lin, Xiaochun Mai
The paper highlights hidden data challenges in clinical machine learning, focusing on how commonly reused preprocessing pipelines for electronic health records embed assumptions about data availability, episode definitions, and missingness. Using MIMIC-IV, the authors identify violations in temporal, semantic, and population alignment—leading to temporal leakage, feature-task misalignment, and biased imputation—and show that these ambiguities can alter model signals without affecting overall predictive performance. They provide practical recommendations and open-source code for more temporally valid, clinically aligned, and reproducible preprocessing.
By Carolin Ciss\'ee, Dennis Fast, Raneen Younis, Alexander L\"oser, Wolfgang Nejdl, Zahra Ahmadi
The paper introduces a feature selection framework for high‑dimensional financial regression that uses Monte Carlo Tree Search with Generalized Nested Rollout Policy Adaptation. It treats feature selection as a sequential decision process and evaluates candidate subsets by out‑of‑sample validation performance, incorporating subset size and multicollinearity into a lexicographic objective. Applied to a NASDAQ‑100 weekly return regression with 1,072 engineered features, the method outperforms an optimized Lasso baseline, achieving an out‑of‑sample R² of 0.6961 versus 0.6313.
By Cameron Mouangue, Tristan Cazenave
CLEAT (CLarification through valuE-guided Agent co-Training) is a framework that jointly trains a clarification-value controller and a task policy to improve interactive intent alignment in large language model agents. It introduces a residual query value metric, estimated via paired rollouts, to assess the utility of asking clarification questions and uses this to guide both controller predictions and policy decisions. Experiments with Qwen3-4B show CLEAT outperforming baselines across multiple benchmarks, with significant gains in user‑centric tasks and demonstrating that co‑training benefits the task policy even when the controller is omitted at inference.
By Jiatan Huang, Zheyuan Zhang, Ziming Li, Tianyi Ma, Mingchen Li, Yanfang Ye, Chuxu Zhang
VEPA is a framework that enables the verified evolution of formal guardrails for lifelong large language model agents. It represents guardrails as formal policies of symbolic rules over typed agent events, and during deployment uses LLM‑guided feedback to propose policy patches that fix new failures while preserving correct behavior. A deterministic verifier accepts only structurally valid, conflict‑free policies, and an evidence‑aware authority layer activates verified rules only when sufficient supporting evidence exists, ensuring that uncertain updates do not directly affect guardrail decisions.
By Duy Nguyen, Lesly Miculicich, Bhavana Dalvi Mishra, Shuo Li, Krishnamurthy Dj Dvijotham, Mohit Bansal, Tomas Pfister, Long T. Le
The paper presents a local agentic AI framework that uses laptop-based language models to autonomously generate and refine edge case scenarios for autonomous systems. It employs a think‑act‑observe‑update loop with JSON encoding, contextual prompts from system requirements and environment descriptions, and a second LM that critiques scenarios against domain criteria. The refined scenarios are simulated, evaluated, and clustered to identify functional gaps, providing traceable evidence for safety assurance.
By Victoria J. Hodge, Colin Paterson, Rob Alexander
The paper presents an unsupervised multimodal reward model built on a frozen Qwen3‑VL backbone. By adding a linear head to an intermediate layer and training it to rank a web caption higher than a perturbed version of itself, the authors avoid human preference labels and achieve competitive performance—outperforming GPT‑4o on several benchmarks with far fewer parameters and compute. The approach demonstrates that a carefully chosen intermediate layer, simple perturbation-based training, and verifier‑driven fine‑tuning can yield effective vision‑language rewards without costly annotation.
By Yueying Li, Jingxuan Fan, Kiant\'e Brantley, Sham M. Kakade, Hanlin Zhang
LeWAM is an end‑to‑end joint‑embedding predictive architecture that learns compact latent world models without pixel reconstruction and directly supports policy execution and planning. It trains jointly on next‑latent‑state prediction and action‑flow matching while imposing an isotropic Gaussian constraint on the latent state. The approach yields significant performance gains, raising visual‑planning success rates from 28.6 % to 89.7 % on contact‑rich tasks, from 10.9 % to 46.2 % on long‑horizon tasks, and speeding up planning by 32.3×, while remaining competitive on behavior‑cloning tasks and providing interpretable control‑relevant representations.
By Minghao Fu, Tavis Siebert, Eryk Halicki, Randall Balestriero
The paper introduces a principled framework for selecting multiple hidden-state layers from large language model decoders to improve speculative decoding. It uses matrix‑entropy to gauge informativeness and centered kernel alignment (CKA) to assess diversity, revealing distinct geometric and informational compartments across layers. The authors propose entropy‑ and CKA‑based selection algorithms, token‑entropy filtering, and demonstrate that these choices outperform traditional fixed‑layer heuristics in EAGLE3‑style decoding.
By Raghavv Goel, Sudhanshu Agrawal, Mukul Gagrani, Risheek Garrepalli, Mingu Lee
The paper introduces RobustBench‑TC, a benchmark that tests tool‑use language agents under 22 real‑world perturbations affecting observations, actions, rewards, and transitions. Experiments on 21 models show that observation noise has minimal impact, while reward‑relevant and transition noise drastically reduce accuracy, and model size alone does not mitigate these gaps. The authors propose ToolRL‑DR, a domain‑randomized reinforcement‑learning recipe that improves robustness, retaining most clean accuracy and narrowing performance gaps even on unseen transition perturbations.
By Xiaolin Zhou, Aojie Yuan, Zheng Luo, Zipeng Ling, Xixiao Pan, Yicheng Gao, Haiyue Zhang, Jiate Li, Shuli Jiang, Prince Zizhuang Wang, Zixuan Zhu, Jinbo Liu, Ryan A. Rossi, Hua Wei, Xiyang Hu
The paper revisits feature inversion via inverse networks, proposing a modular variant that is computationally efficient and can produce more semantically coherent image reconstructions. It applies this method to large‑scale transformer‑based vision models—Detection Transformer, Vision Transformer, Swin Transformer, and Data‑Efficient Image Transformer—examining reconstructions across network depth. The analysis highlights gradual representational changes in Detection and Vision Transformers, systematic differences in preserving contextual shape and fine‑grained details, and varying robustness to color perturbations, while also demonstrating the framework’s broader applicability to other transformer architectures.
By Jan Rathjens, Shirin Reyhanian, David Kappel, Laurenz Wiskott
The paper introduces C$^2$TU, a training‑free and prompt‑free method for uncovering Chinese cloaked toxic content. It uses substring matching with a Chinese homophone graph and toxic lexicon to spot candidate toxic words, then filters non‑toxic candidates and corrects cloaks using two variants—one based on BERT and another on LLMs that overcome auto‑regressive limitations. Experiments on two Chinese toxic datasets show C$^2$TU outperforms the best competitor by up to 71% in F1 score and 35% in accuracy.
By Xuchen Ma, Jianxiang Yu, Wenming Shao, Bo Pang, Xiang Li
UniConFlow is a unified constrained flow‑matching framework that generates robot trajectories while systematically handling both equality and inequality constraints such as collision avoidance, actuation limits, and dynamic consistency. It introduces a prescribed‑time zeroing function to shape a time‑varying guidance field during inference, and employs a violation‑segment extraction protocol and trajectory compression to address computational challenges in long‑horizon, high‑dimensional planning. Experiments on a double inverted pendulum, a real‑to‑sim car racing task, and a sim‑to‑real manipulation task show that UniConFlow outperforms state‑of‑the‑art generative planners and conventional optimization baselines on certified motion planning metrics like safety, kinodynamic consistency, and action feasibility.
By Zewen Yang, Xiaobing Dai, Dian Yu, Zhijun Li, Majid Khadiv, Sandra Hirche, Sami Haddadin
LLMail-Inject is a large-scale benchmark comprising 208,095 human-written prompt injection attacks gathered from a public red‑teaming challenge with 839 participants. The dataset tests the full attack chain—retrieving prompts from an email database, bypassing production defenses, and triggering unauthorized tool calls with correct arguments. Evaluation shows that models performing well on smaller benchmarks struggle on LLMail-Inject, with GPT‑4o achieving 58% attack success and Meta SecAlign 32%.
By Sahar Abdelnabi, Aideen Fay, Ahmed Salem, Egor Zverev, Kai-Chieh Liao, Chi-Huang Liu, Chun-Chih Kuo, Jannis Weigend, Danyael Manlangit, Alex Apostolov, Haris Umair, Jo\~ao Donato, Masayuki Kawakita, Athar Mahboob, Tran Huu Bach, Tsun-Han Chiang, Myeongjin Cho, Hajin Choi, Byeonghyeon Kim, Hyeonjin Lee, Benjamin Pannell, Conor McCauley, Mark Russinovich, Andrew Paverd, Giovanni Cherubin
The paper proposes a model‑free, instance‑based learning framework that uses surprisal—an information‑theoretic measure of uncertainty—to analyze raw data directly, bypassing explicit distribution modeling. By quantifying relevance through uncertainty, the method supports generalizable inference across diverse tasks such as generative modeling, causal discovery, anomaly detection, and time‑series forecasting, while offering traceability and interpretability. The authors claim state‑of‑the‑art performance on anomaly detection and some supervised tasks, and argue that the mathematical foundations create a ‘physics’ of information that can underpin scalable, human‑understandable AI.
By Christopher J. Hazard, Michael Resnick, Jacob Beel, Jack Xia, Cade Mack, Dominic Glennie, Matthew Fulp, David Maze, Andrew Bassett, Martin Koistinen
The paper analyzes Gradient Descent (GD) on a minimal binary classification problem with a two‑neuron ReLU network trained on two instances. It proves that GD converges to an optimal robustness margin—maximizing the distance between the decision boundary and training points—but does so at a very slow rate, specifically θ(1/ln(t)). Empirical simulations confirm this slow convergence across various initializations, and the authors provide a rigorous analysis of GD trajectories across different activation patterns to establish the first explicit lower bound on robustness convergence in a non‑linear model.
By Guy Smorodinsky, Sveta Gimpleson, Itay Safran
The paper discusses how the shift from centralized AI APIs to local edge inference enables Client‑Side Autonomous Agents (CSAAs) that collaborate in Agentic Peer‑to‑Peer (P2P) networks. Unlike traditional P2P systems that exchange static content, these networks share dynamic, heterogeneous capabilities that may be unsafe, necessitating semantic discovery, drift management, and risk‑aware execution. The authors propose a plane‑based reference architecture, signed soft‑state capability descriptors for intent‑aware routing, and a tiered verification spectrum (reputation, canary probes, cryptographic evidence) to align trust costs with task risks, showing improved workflow success in simulations.
By Taotao Wang, Lizhao You, Jingwen Tong, Chonghe Zhao, Shengli Zhang
MyoSem is an EMG–action semantic alignment framework that maps hand electromyography signals and multi‑view natural‑language action descriptions into a shared feature space for bidirectional retrieval. It constructs semantic prototypes from text, encodes EMG with local activation‑aware representations, and aggregates evidence using semantic queries. Evaluated on multiple datasets and protocols, MyoSem outperforms existing methods in bidirectional EMG‑text retrieval, demonstrating its effectiveness in moving EMG hand action understanding toward language‑queryable semantic retrieval.
By Chiyue Wang, Yang Gao, Zhanpeng Jin
The paper introduces a multi-agent peer‑reviewed reasoning approach for medical question answering, where several large language models independently generate chain‑of‑thought explanations and then evaluate each other’s reasoning for factual correctness and logical soundness. The highest‑rated reasoning chain is used to produce the final answer. Experiments with five state‑of‑the‑art LLMs on three benchmark datasets show that this method consistently outperforms single‑model chain‑of‑thought reasoning and majority voting ensembles, achieving an average accuracy of 0.820 versus 0.777 for the best single model.
By Zaifu Zhan, Shuang Zhou, Rui Zhang
The paper introduces First-Order In-Context Shapley, an inference‑only method that evaluates data utility for large language model alignment by computing the first‑order term of an in‑context cooperative game over semantic k‑NN neighborhoods in O(k) forward passes. It projects log‑likelihood shifts onto a directed influence graph, yielding three signals—Exemplifying Advantage, Exemplified Advantage, and One‑of‑a‑Kind Score—to identify high‑utility demonstrations and rare prompts. The approach reduces compute by 19× compared to Monte‑Carlo Shapley, filters 1–5% of records, and improves error detection and reward accuracy in DPO training and benchmark auditing.
whyItMatters":"By enabling scalable, inference‑only data auditing, the method addresses label noise and safety inversions that hinder post‑training alignment of large language models, leading to measurable improvements in model reliability and reward accuracy."
By Yunting Song, Matthew Watson, Moustafa Farid Alzantot, Peter Grabowski, Jun Qin