We consider representation misdirection (RM), a class of LLM unlearning methods that achieves forgetting by manipulating the forget-representations, that is, latent representations of forget samples. Despite being important, the roles of target vectors used in RM, however, remain underexplored. Here, we approach and revisit RM through the lens of the linear representation hypothesis. Specifically, if one can somehow identify a one-dimensional representation corresponding to a high-level concept, the linear representation hypothesis enables linear operations on this concept vector […]

Ver mais

How Pattern Recognition Architecture Accidentally Became Behavioral Psychology at Scale The discourse surrounding artificial intelligence has long centered on computational capability — model parameters, benchmark scores, reasoning depth. Yet the most profound transformation in human-AI interaction stems not from architectural sophistication, but from an emergent capability that was never explicitly programmed: affective pattern recognition at the micro-behavioral level. What we’re witnessing isn’t the creation of artificial empathy. It’s something far more consequential: the systematic extraction and modeling of human emotional architecture […]

Ver mais

This paper presents a geology-driven machine learning method for automated rock joint trace mapping from images. The approach combines geological modelling, synthetic data generation, and supervised image segmentation to address limited real data and class imbalance. First, discrete fracture network models are used to generate synthetic jointed rock images at field-relevant scales via parametric modelling, preserving joint persistence, connectivity, and node-type distributions. Second, segmentation models are trained using mixed training and pretraining followed by fine-tuning on real images. […]

Ver mais

arXiv:2602.07453v1 Announce Type: cross Abstract: Decision tree ensembles are widely used in critical domains, making robustness and sensitivity analysis essential to their trustworthiness. We study the feature sensitivity problem, which asks whether an ensemble is sensitive to a specified subset of features — such as protected attributes — whose manipulation can alter model predictions. Existing approaches often yield examples of sensitivity that lie far from the training distribution, limiting their interpretability and practical value. We propose a data-aware […]

Ver mais

arXiv:2602.17827v1 Announce Type: cross Abstract: Generative Flow Networks (GFlowNets) are a flexible family of amortized samplers trained to generate discrete and compositional objects with probability proportional to a reward function. However, learning efficiency is constrained by the model’s ability to rapidly explore diverse high-probability regions during training. To mitigate this issue, recent works have focused on incentivizing the exploration of unvisited and valuable states via curiosity-driven search and self-supervised random network distillation, which tend to waste samples on […]

Ver mais

[On agents using CLI tools in place of REST APIs] To save on context window, yes, but moreso to improve accuracy and success rate when multiple tool calls are involved, particularly when calls must be correctly chained e.g. for pagination, rate-limit backoff, and recognizing authentication failures. Other major factor: which models can wield the skill? Using the CLI lowers the bar so cheap, fast models (gpt-5-nano, haiku-4.5) can reliably succeed. Using the raw APl is something only the […]

Ver mais

In this article I want to explain how a Principal Component Analysis (PCA) works by implementing it in Python step by step. At the end we will compare the results to the more convenient Python PCA() classes that are available through the popular matplotlib and scipy libraries and discuss how they differ.

Ver mais

arXiv:2602.10545v1 Announce Type: cross Abstract: Modern large-scale neural networks are often trained and released in multiple sizes to accommodate diverse inference budgets. To improve efficiency, recent work has explored model upscaling: initializing larger models from trained smaller ones in order to transfer knowledge and accelerate convergence. However, this method can be sensitive to hyperparameters that need to be tuned at the target upscaled model size, which is prohibitively costly to do directly. It remains unclear whether the most […]

Ver mais

arXiv:2601.14495v1 Announce Type: new Abstract: There are several paradigms for integrating interactive and automated theorem provers, combining the convenience of powerful automation with strong soundness guarantees. We introduce a new approach for reconstructing proofs found by SMT solvers which we intend to be complementary with existing techniques. Rather than verifying or replaying a full proof produced by the SMT solver, or at the other extreme, rediscovering the solver’s proof from just the set of premises it uses, we […]

Ver mais