How to Enrich LLM Context to Significantly Enhance CapabilitiesTowards Data Scienceon September 16, 2025 at 2:00 pm Learn how to empower your LLMs by leveraging additional metadata
The post How to Enrich LLM Context to Significantly Enhance Capabilities appeared first on Towards Data Science.
Learn how to empower your LLMs by leveraging additional metadata
The post How to Enrich LLM Context to Significantly Enhance Capabilities appeared first on Towards Data Science. Read More
Kalman Bayesian Transformercs.AI updates on arXiv.orgon September 16, 2025 at 4:00 am arXiv:2509.10695v1 Announce Type: cross
Abstract: Sequential fine-tuning of transformers is useful when new data arrive sequentially, especially with shifting distributions. Unlike batch learning, sequential learning demands that training be stabilized despite a small amount of data by balancing new information and previously learned knowledge in the pre-trained models. This challenge is further complicated when training is to be completed in latency-critical environments and learning must additionally quantify and be mediated by uncertainty. Motivated by these challenges, we propose a novel method that frames sequential fine-tuning as a posterior inference problem within a Bayesian framework. Our approach integrates closed-form moment propagation of random variables, Kalman Bayesian Neural Networks, and Taylor approximations of the moments of softmax functions. By explicitly accounting for pre-trained models as priors and adaptively balancing them against new information based on quantified uncertainty, our method achieves robust and data-efficient sequential learning. The effectiveness of our method is demonstrated through numerical simulations involving sequential adaptation of a decision transformer to tasks characterized by distribution shifts and limited memory resources.
arXiv:2509.10695v1 Announce Type: cross
Abstract: Sequential fine-tuning of transformers is useful when new data arrive sequentially, especially with shifting distributions. Unlike batch learning, sequential learning demands that training be stabilized despite a small amount of data by balancing new information and previously learned knowledge in the pre-trained models. This challenge is further complicated when training is to be completed in latency-critical environments and learning must additionally quantify and be mediated by uncertainty. Motivated by these challenges, we propose a novel method that frames sequential fine-tuning as a posterior inference problem within a Bayesian framework. Our approach integrates closed-form moment propagation of random variables, Kalman Bayesian Neural Networks, and Taylor approximations of the moments of softmax functions. By explicitly accounting for pre-trained models as priors and adaptively balancing them against new information based on quantified uncertainty, our method achieves robust and data-efficient sequential learning. The effectiveness of our method is demonstrated through numerical simulations involving sequential adaptation of a decision transformer to tasks characterized by distribution shifts and limited memory resources. Read More
Why Your A/B Test Winner Might Just Be Random NoiseTowards Data Scienceon September 16, 2025 at 12:30 pm What a coach’s warm-up trial can teach us about running better experiments
The post Why Your A/B Test Winner Might Just Be Random Noise appeared first on Towards Data Science.
What a coach’s warm-up trial can teach us about running better experiments
The post Why Your A/B Test Winner Might Just Be Random Noise appeared first on Towards Data Science. Read More
First RAG, Second SEG: A Training-Free Paradigm for Camouflaged Object Detectioncs.AI updates on arXiv.orgon September 16, 2025 at 4:00 am arXiv:2508.15313v2 Announce Type: replace-cross
Abstract: Camouflaged object detection (COD) poses a significant challenge in computer vision due to the high similarity between objects and their backgrounds. Existing approaches often rely on heavy training and large computational resources. While foundation models such as the Segment Anything Model (SAM) offer strong generalization, they still struggle to handle COD tasks without fine-tuning and require high-quality prompts to yield good performance. However, generating such prompts manually is costly and inefficient. To address these challenges, we propose textbf{First RAG, Second SEG (RAG-SEG)}, a training-free paradigm that decouples COD into two stages: Retrieval-Augmented Generation (RAG) for generating coarse masks as prompts, followed by SAM-based segmentation (SEG) for refinement. RAG-SEG constructs a compact retrieval database via unsupervised clustering, enabling fast and effective feature retrieval. During inference, the retrieved features produce pseudo-labels that guide precise mask generation using SAM2. Our method eliminates the need for conventional training while maintaining competitive performance. Extensive experiments on benchmark COD datasets demonstrate that RAG-SEG performs on par with or surpasses state-of-the-art methods. Notably, all experiments are conducted on a textbf{personal laptop}, highlighting the computational efficiency and practicality of our approach. We present further analysis in the Appendix, covering limitations, salient object detection extension, and possible improvements. textcolor{blue} {Code: https://github.com/Lwt-diamond/RAG-SEG.}
arXiv:2508.15313v2 Announce Type: replace-cross
Abstract: Camouflaged object detection (COD) poses a significant challenge in computer vision due to the high similarity between objects and their backgrounds. Existing approaches often rely on heavy training and large computational resources. While foundation models such as the Segment Anything Model (SAM) offer strong generalization, they still struggle to handle COD tasks without fine-tuning and require high-quality prompts to yield good performance. However, generating such prompts manually is costly and inefficient. To address these challenges, we propose textbf{First RAG, Second SEG (RAG-SEG)}, a training-free paradigm that decouples COD into two stages: Retrieval-Augmented Generation (RAG) for generating coarse masks as prompts, followed by SAM-based segmentation (SEG) for refinement. RAG-SEG constructs a compact retrieval database via unsupervised clustering, enabling fast and effective feature retrieval. During inference, the retrieved features produce pseudo-labels that guide precise mask generation using SAM2. Our method eliminates the need for conventional training while maintaining competitive performance. Extensive experiments on benchmark COD datasets demonstrate that RAG-SEG performs on par with or surpasses state-of-the-art methods. Notably, all experiments are conducted on a textbf{personal laptop}, highlighting the computational efficiency and practicality of our approach. We present further analysis in the Appendix, covering limitations, salient object detection extension, and possible improvements. textcolor{blue} {Code: https://github.com/Lwt-diamond/RAG-SEG.} Read More
The looming crackdown on AI companionshipMIT Technology Reviewon September 16, 2025 at 9:00 am As long as there has been AI, there have been people sounding alarms about what it might do to us: rogue superintelligence, mass unemployment, or environmental ruin from data center sprawl. But this week showed that another threat entirely—that of kids forming unhealthy bonds with AI—is the one pulling AI safety out of the academic…
As long as there has been AI, there have been people sounding alarms about what it might do to us: rogue superintelligence, mass unemployment, or environmental ruin from data center sprawl. But this week showed that another threat entirely—that of kids forming unhealthy bonds with AI—is the one pulling AI safety out of the academic… Read More
Bridging Cultural Distance Between Models Default and Local Classroom Demands: How Global Teachers Adopt GenAI to Support Everyday Teaching Practicescs.AI updates on arXiv.orgon September 16, 2025 at 4:00 am arXiv:2509.10780v1 Announce Type: cross
Abstract: Generative AI (GenAI) is rapidly entering K-12 classrooms, offering teachers new ways for teaching practices. Yet GenAI models are often trained on culturally uneven datasets, embedding a “default culture” that often misaligns with local classrooms. To understand how teachers navigate this gap, we defined the new concept Cultural Distance (the gap between GenAI’s default cultural repertoire and the situated demands of teaching practice) and conducted in-depth interviews with 30 K-12 teachers, 10 each from South Africa, Taiwan, and the United States, who had integrated AI into their teaching practice. These teachers’ experiences informed the development of our three-level cultural distance framework. This work contributes the concept and framework of cultural distance, six illustrative instances spanning in low, mid, high distance levels with teachers’ experiences and strategies for addressing them. Empirically, we offer implications to help AI designers, policymakers, and educators create more equitable and culturally responsive GenAI tools for education.
arXiv:2509.10780v1 Announce Type: cross
Abstract: Generative AI (GenAI) is rapidly entering K-12 classrooms, offering teachers new ways for teaching practices. Yet GenAI models are often trained on culturally uneven datasets, embedding a “default culture” that often misaligns with local classrooms. To understand how teachers navigate this gap, we defined the new concept Cultural Distance (the gap between GenAI’s default cultural repertoire and the situated demands of teaching practice) and conducted in-depth interviews with 30 K-12 teachers, 10 each from South Africa, Taiwan, and the United States, who had integrated AI into their teaching practice. These teachers’ experiences informed the development of our three-level cultural distance framework. This work contributes the concept and framework of cultural distance, six illustrative instances spanning in low, mid, high distance levels with teachers’ experiences and strategies for addressing them. Empirically, we offer implications to help AI designers, policymakers, and educators create more equitable and culturally responsive GenAI tools for education. Read More
Tabular Data with Class Imbalance: Predicting Electric Vehicle Crash Severity with Pretrained Transformers (TabPFN) and Mamba-Based Modelscs.AI updates on arXiv.orgon September 16, 2025 at 4:00 am arXiv:2509.11449v1 Announce Type: cross
Abstract: This study presents a deep tabular learning framework for predicting crash severity in electric vehicle (EV) collisions using real-world crash data from Texas (2017-2023). After filtering for electric-only vehicles, 23,301 EV-involved crash records were analyzed. Feature importance techniques using XGBoost and Random Forest identified intersection relation, first harmful event, person age, crash speed limit, and day of week as the top predictors, along with advanced safety features like automatic emergency braking. To address class imbalance, Synthetic Minority Over-sampling Technique and Edited Nearest Neighbors (SMOTEENN) resampling was applied. Three state-of-the-art deep tabular models, TabPFN, MambaNet, and MambaAttention, were benchmarked for severity prediction. While TabPFN demonstrated strong generalization, MambaAttention achieved superior performance in classifying severe injury cases due to its attention-based feature reweighting. The findings highlight the potential of deep tabular architectures for improving crash severity prediction and enabling data-driven safety interventions in EV crash contexts.
arXiv:2509.11449v1 Announce Type: cross
Abstract: This study presents a deep tabular learning framework for predicting crash severity in electric vehicle (EV) collisions using real-world crash data from Texas (2017-2023). After filtering for electric-only vehicles, 23,301 EV-involved crash records were analyzed. Feature importance techniques using XGBoost and Random Forest identified intersection relation, first harmful event, person age, crash speed limit, and day of week as the top predictors, along with advanced safety features like automatic emergency braking. To address class imbalance, Synthetic Minority Over-sampling Technique and Edited Nearest Neighbors (SMOTEENN) resampling was applied. Three state-of-the-art deep tabular models, TabPFN, MambaNet, and MambaAttention, were benchmarked for severity prediction. While TabPFN demonstrated strong generalization, MambaAttention achieved superior performance in classifying severe injury cases due to its attention-based feature reweighting. The findings highlight the potential of deep tabular architectures for improving crash severity prediction and enabling data-driven safety interventions in EV crash contexts. Read More
Privacy-Preserving Decentralized Federated Learning via Explainable Adaptive Differential Privacycs.AI updates on arXiv.orgon September 16, 2025 at 4:00 am arXiv:2509.10691v1 Announce Type: cross
Abstract: Decentralized federated learning faces privacy risks because model updates can leak data through inference attacks and membership inference, a concern that grows over many client exchanges. Differential privacy offers principled protection by injecting calibrated noise so confidential information remains secure on resource-limited IoT devices. Yet without transparency, black-box training cannot track noise already injected by previous clients and rounds, which forces worst-case additions and harms accuracy. We propose PrivateDFL, an explainable framework that joins hyperdimensional computing with differential privacy and keeps an auditable account of cumulative noise so each client adds only the difference between the required noise and what has already been accumulated. We evaluate on MNIST, ISOLET, and UCI-HAR to span image, signal, and tabular modalities, and we benchmark against transformer-based and deep learning-based baselines trained centrally with Differentially Private Stochastic Gradient Descent (DP-SGD) and Renyi Differential Privacy (RDP). PrivateDFL delivers higher accuracy, lower latency, and lower energy across IID and non-IID partitions while preserving formal (epsilon, delta) guarantees and operating without a central server. For example, under non-IID partitions, PrivateDFL achieves 24.42% higher accuracy than the Vision Transformer on MNIST while using about 10x less training time, 76x lower inference latency, and 11x less energy, and on ISOLET it exceeds Transformer accuracy by more than 80% with roughly 10x less training time, 40x lower inference latency, and 36x less training energy. Future work will extend the explainable accounting to adversarial clients and adaptive topologies with heterogeneous privacy budgets.
arXiv:2509.10691v1 Announce Type: cross
Abstract: Decentralized federated learning faces privacy risks because model updates can leak data through inference attacks and membership inference, a concern that grows over many client exchanges. Differential privacy offers principled protection by injecting calibrated noise so confidential information remains secure on resource-limited IoT devices. Yet without transparency, black-box training cannot track noise already injected by previous clients and rounds, which forces worst-case additions and harms accuracy. We propose PrivateDFL, an explainable framework that joins hyperdimensional computing with differential privacy and keeps an auditable account of cumulative noise so each client adds only the difference between the required noise and what has already been accumulated. We evaluate on MNIST, ISOLET, and UCI-HAR to span image, signal, and tabular modalities, and we benchmark against transformer-based and deep learning-based baselines trained centrally with Differentially Private Stochastic Gradient Descent (DP-SGD) and Renyi Differential Privacy (RDP). PrivateDFL delivers higher accuracy, lower latency, and lower energy across IID and non-IID partitions while preserving formal (epsilon, delta) guarantees and operating without a central server. For example, under non-IID partitions, PrivateDFL achieves 24.42% higher accuracy than the Vision Transformer on MNIST while using about 10x less training time, 76x lower inference latency, and 11x less energy, and on ISOLET it exceeds Transformer accuracy by more than 80% with roughly 10x less training time, 40x lower inference latency, and 36x less training energy. Future work will extend the explainable accounting to adversarial clients and adaptive topologies with heterogeneous privacy budgets. Read More
An Entropy-Guided Curriculum Learning Strategy for Data-Efficient Acoustic Scene Classification under Domain Shiftcs.AI updates on arXiv.orgon September 16, 2025 at 4:00 am arXiv:2509.11168v1 Announce Type: cross
Abstract: Acoustic Scene Classification (ASC) faces challenges in generalizing across recording devices, particularly when labeled data is limited. The DCASE 2024 Challenge Task 1 highlights this issue by requiring models to learn from small labeled subsets recorded on a few devices. These models need to then generalize to recordings from previously unseen devices under strict complexity constraints. While techniques such as data augmentation and the use of pre-trained models are well-established for improving model generalization, optimizing the training strategy represents a complementary yet less-explored path that introduces no additional architectural complexity or inference overhead. Among various training strategies, curriculum learning offers a promising paradigm by structuring the learning process from easier to harder examples. In this work, we propose an entropy-guided curriculum learning strategy to address the domain shift problem in data-efficient ASC. Specifically, we quantify the uncertainty of device domain predictions for each training sample by computing the Shannon entropy of the device posterior probabilities estimated by an auxiliary domain classifier. Using entropy as a proxy for domain invariance, the curriculum begins with high-entropy samples and gradually incorporates low-entropy, domain-specific ones to facilitate the learning of generalizable representations. Experimental results on multiple DCASE 2024 ASC baselines demonstrate that our strategy effectively mitigates domain shift, particularly under limited labeled data conditions. Our strategy is architecture-agnostic and introduces no additional inference cost, making it easily integrable into existing ASC baselines and offering a practical solution to domain shift.
arXiv:2509.11168v1 Announce Type: cross
Abstract: Acoustic Scene Classification (ASC) faces challenges in generalizing across recording devices, particularly when labeled data is limited. The DCASE 2024 Challenge Task 1 highlights this issue by requiring models to learn from small labeled subsets recorded on a few devices. These models need to then generalize to recordings from previously unseen devices under strict complexity constraints. While techniques such as data augmentation and the use of pre-trained models are well-established for improving model generalization, optimizing the training strategy represents a complementary yet less-explored path that introduces no additional architectural complexity or inference overhead. Among various training strategies, curriculum learning offers a promising paradigm by structuring the learning process from easier to harder examples. In this work, we propose an entropy-guided curriculum learning strategy to address the domain shift problem in data-efficient ASC. Specifically, we quantify the uncertainty of device domain predictions for each training sample by computing the Shannon entropy of the device posterior probabilities estimated by an auxiliary domain classifier. Using entropy as a proxy for domain invariance, the curriculum begins with high-entropy samples and gradually incorporates low-entropy, domain-specific ones to facilitate the learning of generalizable representations. Experimental results on multiple DCASE 2024 ASC baselines demonstrate that our strategy effectively mitigates domain shift, particularly under limited labeled data conditions. Our strategy is architecture-agnostic and introduces no additional inference cost, making it easily integrable into existing ASC baselines and offering a practical solution to domain shift. Read More
A Visual Guide to Tuning Gradient Boosted TreesTowards Data Scienceon September 15, 2025 at 6:59 pm Introduction My previous posts looked at the bog-standard decision tree and the wonder of a random forest. Now, to complete the triplet, I’ll visually explore gradient boosted trees! There are a bunch of gradient boosted tree libraries, including XGBoost, CatBoost, and LightGBM. However, for this I’m going to use sklearn’s one. Why? Simply because, compared
The post A Visual Guide to Tuning Gradient Boosted Trees appeared first on Towards Data Science.
Introduction My previous posts looked at the bog-standard decision tree and the wonder of a random forest. Now, to complete the triplet, I’ll visually explore gradient boosted trees! There are a bunch of gradient boosted tree libraries, including XGBoost, CatBoost, and LightGBM. However, for this I’m going to use sklearn’s one. Why? Simply because, compared
The post A Visual Guide to Tuning Gradient Boosted Trees appeared first on Towards Data Science. Read More