With the goal of understanding the visual concepts that CLIP associates with text prompts, we show that the latent space of CLIP can be visualized solely in terms of linear transformations on simple geometric primitives like straight lines and circles. Although existing approaches achieve this by sketch-synthesisthrough-optimization, they do so on the space of higher order Bézier curves, which exhibit a wastefully large set of structures that they can evolve into, as most of them are non-essential for generating meaningful sketches. We present CLIPDraw++, an algorithm that provides significantly better visualizations for CLIP text embeddings, using only simple primitive shapes like straight lines and circles. This constrains the set of possible outputs to linear transformations on these primitives, thereby exhibiting an inherently simpler mathematical form. The synthesis process of CLIPDraw++ can be tracked end-to-end, with each visual concept being expressed exclusively in terms of primitives.

Recent progress in artificial intelligence has been driven by empirical scaling laws linking performance improvements to increases in model parameters, data, and compute. These results have fueled the widespread belief that artificial general intelligence (AGI) will emerge primarily through continued scaling of large language models (LLMs). In this paper, we argue that this assumption conflates benchmark performance with intelligence and overlooks fundamental architectural limitations of current models. We propose Artificial Special Intelligence (ASI) as an alternative framework: intelligence arising from collections of small, specialized models that operate asynchronously, learn continuously, and interact with large-scale external memory. Drawing on evidence from machine learning, neuroscience, and cognitive science, we argue that intelligence is better characterized by structural properties—such as specialization, separation of compute and memory, and lifelong learning than by parameter count alone.

Neural text-to-speech systems systematically mispronounce low-resource proper nouns, particularly non-English names, brands, and geographic locations due to their underrepresentation in predominantly English training corpora. Existing solutions require expensive multilingual data collection or manual phonetic annotation, limiting TTS deployment in diverse linguistic contexts. We introduce SonoEdit, a model editing technique that surgically corrects pronunciation errors in pre-trained TTS models without retraining. Correcting such errors traditionally requires costly supervised finetuning or manual phoneme injection. In this work, we present a parsimonious alternative using Null-Space Pronunciation Editing, a single-shot parameter update that modifies the pronunciation of specific words while provably preserving the rest of the model's behavior. We first adapt Acoustic Causal Tracing to identify the specific Transformer layers governing text-to-pronunciation mapping. We then employ Null-Space Constrained Editing to compute a closed-form weight update that rectifies the target pronunciation while remaining mathematically orthogonal to the manifold of general speech, constructing a constrained update that drives the model's acoustic output toward a desired pronunciation exemplar while ensuring zero first-order change on a preserved corpus.

Third Conference on Parsimony and Learning (CPAL 2026).

Abstract

We develop a robust translation model for four low-resource Indic languages: Khasi, Mizo, Manipuri, and Assamese. Our approach includes a comprehensive pipeline from data collection and preprocessing to training and evaluation, leveraging data from WMT task datasets, BPCC, PMIndia, and OpenLanguageData. To address the scarcity of bilingual data, we use back-translation techniques on monolingual datasets for Mizo and Khasi, significantly expanding our training corpus. We fine-tune the pre-trained NLLB 3.3B model for Assamese, Mizo, and Manipuri, achieving improved performance over the baseline. For Khasi, which is not supported by the NLLB model, we introduce special tokens and train the model on our Khasi corpus. Our training involves masked language modelling, followed by fine-tuning for English-to-Indic and Indic-to-English translations.