arXiv:2603.26778v1 Announce Type: new Abstract: Knowledge distillation is typically realized by transferring a teacher model’s knowledge into a student’s parameters through supervised or reinforcement-based optimization. While effective, such approaches require repeated parameter updates and large-scale training data, limiting their applicability in resource-constrained environments. In this work, we propose TED, a training-free, context-based distillation framework that shifts the update target of distillation from model parameters to an in-context experience injected into the student’s prompt. For each input, the student […]

Ver mais

Hi, I’ve created an overview of the most important OCR benchmarks, along with the top open models, and links to their paper and code: https://paperswithcode.co/tasks/ocr. This week, new OCR models were released by Baidu and Mistral. Baidu released Unlimited OCR, a 3B-parameter model that introduces a key innovation called Reference Sliding Window Attention (R-SWA) and builds on top of DeepSeek OCR. Mistral released OCR 4, which is available via an API. OCR, or Optical-Character Recognition, is the task […]

Ver mais

arXiv:2601.08986v1 Announce Type: new Abstract: We study the pointwise maximal leakage (PML) envelope of the Gaussian mechanism, which characterizes the smallest information leakage bound that holds with high probability under arbitrary post-processing. For the Gaussian mechanism with a Gaussian secret, we derive a closed-form expression for the deterministic PML envelope for sufficiently small failure probabilities. We then extend this result to general unbounded secrets by identifying a sufficient condition under which the envelope coincides with the Gaussian case. […]

Ver mais

Introduction Disclaimer: This article is a cross post from Pytorch Medium Blog Post. One of the most exciting advancements, that has pushed the frontier of the Artificial Intelligence (AI) in recent years, is Deep Reinforcement Learning (DRL). DRL belongs to the family of machine learning algorithms. It assumes that intelligent machines can learn from their actions similar to the way humans learn from experience. Over the recent years we could witness some impressive real-world applications of DRL. The […]

Ver mais

Hi everyone, I’m an independent researcher (incoming MSc AI, University of Edinburgh) and I’ve written a pre-registration paper modelling the 2026 Formula 1 energy regulations as a Partially Observable Stochastic Game. I’m looking for an arXiv endorsement in cs.AI or cs.LG to upload it before the Melbourne GP on 8 March, ideally even before the race weekend starts. The paper: Opponent State Inference Under Partial Observability: An HMM–POMDP Framework for 2026 Formula 1 Energy Strategy https://www.researchgate.net/publication/401368044_Opponent_State_Inference_Under_Partial_Observability_An_HMM-POMDP_Framework_for_2026_Formula_1_Energy_Strategy The problem: […]

Ver mais

Response variability limits the clinical utility of transcutaneous auricular vagus nerve stimulation (taVNS) for negative symptoms in treatment-resistant schizophrenia (TRS). This study aimed to develop an electroencephalography (EEG)-based machine learning (ML) model to predict individual response and explore associated neurophysiological mechanisms. We used ML to develop and validate predictive models based on pre-treatment EEG data features (power, coherence, and dynamic functional connectivity) from 50 TRS patients enrolled in the taVNS trial, within a nested cross-validation framework. Participants received […]

Ver mais

arXiv:2602.15891v1 Announce Type: new Abstract: While autonomous vehicles have achieved reliable performance within specific operating regions, their deployment to new cities remains costly and slow. A key bottleneck is the need to collect many human demonstration trajectories when adapting driving policies to new cities that differ from those seen in training in terms of road geometry, traffic rules, and interaction patterns. In this paper, we show that self-play multi-agent reinforcement learning can adapt a driving policy to a […]

Ver mais