arXiv:2602.03863v1 Announce Type: new Abstract: Computational reproducibility, the possibility for independent researchers to exactly reproduce published empirical results, is fundamental to science. Despite its importance, the proportion of research articles aiming for reproducibility remains low and uneven across disciplines. Barriers include a perceived lack of incentives for researchers and journals, practical challenges in preparing reproducible materials, and the absence of harmonised standards of reproducibility processes and requirements by journals. Existing guidance is often highly technical, reaching mainly those […]

Ver mais

arXiv:2603.02400v1 Announce Type: new Abstract: In this paper, we investigate the impact of the broadcast effect arising in filterless optical networks on the computational complexity of the wavelength assignment problem. We model conflicts using an appropriate interference digraph, whose proper colourings correspond to feasible wavelength assignments. Minimizing the number of required wavelengths therefore amounts to determining the chromatic number of this interference digraph. Within this framework, we first present a polynomial-time 2-approximation algorithm for minimizing the number of […]

Ver mais

arXiv:2602.02591v1 Announce Type: new Abstract: We introduce and define a novel task-Scene-Aware Visually-Driven Speech Synthesis, aimed at addressing the limitations of existing speech generation models in creating immersive auditory experiences that align with the real physical world. To tackle the two core challenges of data scarcity and modality decoupling, we propose VividVoice, a unified generative framework. First, we constructed a large-scale, high-quality hybrid multimodal dataset, Vivid-210K, which, through an innovative programmatic pipeline, establishes a strong correlation between visual […]

Ver mais

Dense image captioning is critical for cross-modal alignment in vision-language pretraining and text-to-image generation, but scaling expert-quality annotations is prohibitively expensive. While synthetic captioning via strong vision-language models (VLMs) is a practical alternative, supervised distillation often yields limited output diversity and weak generalization. Reinforcement learning (RL) could overcome these limitations, but its successes have so far been concentrated in verifiable domains that rely on deterministic checkers — a luxury not available in open-ended captioning. We address this bottleneck […]

Ver mais

arXiv:2603.26763v1 Announce Type: new Abstract: Talking-head videos constitute a predominant content type in real-time communication, yet publicly available datasets for video processing research in this domain remain scarce and limited in signal fidelity. In this paper, we open-source a near-raw dataset of 847 talking-head recordings (approximately 212 minutes), each 15,s in duration, captured from 805 participants using 446 unique consumer webcam devices in their natural environments. All recordings are stored using the FFV1 lossless codec, preserving the camera-native […]

Ver mais

We study the canonical string-based Assembly Index (ASI), defined as the minimum number of binary concatenations needed to construct a target word under full reuse. NP-completeness of ASI-DEC over general finite alphabets and an equivalence between ASI plans and straight-line programs (SLPs) under the same size convention has been established. We emphasize that all transfers between decision variants are effected by explicit polynomial-time mappings and (where needed) an explicit reparameterization of the threshold by an absolute constant or […]

Ver mais

Large language models (LLMs) are increasingly trained in complex Reinforcement Learning, multi-agent environments, making it difficult to understand how behavior changes over training. Sparse Autoencoders (SAEs) have recently shown to be useful for data-centric interpretability. In this work, we analyze large-scale reinforcement learning training runs from the sophisticated environment of Full-Press Diplomacy by applying pretrained SAEs, alongside LLM-summarizer methods. We introduce Meta-Autointerp, a method for grouping SAE features into interpretable hypotheses about training dynamics. We discover fine-grained behaviors […]

Ver mais

arXiv:2602.18568v1 Announce Type: new Abstract: Large language model (LLM) inference performance is increasingly bottlenecked by the memory wall. While GPUs continue to scale raw compute throughput, they struggle to deliver scalable performance for memory bandwidth bound workloads. This challenge is amplified by emerging reasoning LLM applications, where long output sequences, low arithmetic intensity, and tight latency constraints demand significantly higher memory bandwidth. As a result, system utilization drops and energy per inference rises, highlighting the need for an […]

Ver mais

arXiv:2309.08945v2 Announce Type: replace-cross Abstract: In recent years, a certain type of problems have become of interest where one wants to query a trained classifier. Specifically, one wants to find the closest instance to a given input instance such that the classifier’s predicted label is changed in a desired way. Examples of these “inverse classification” problems are counterfactual explanations, adversarial examples and model inversion. All of them are fundamentally optimization problems over the input instance vector involving a […]

Ver mais