DIVE: Scaling Diversity in Agentic Task Synthesis for Generalizable Tool Use
arXiv:2603.11076v1 Announce Type: new Abstract: Recent work synthesizes agentic tasks for post-training tool-using LLMs, yet robust generalization under shifts in…
Historical Consensus: Preventing Posterior Collapse via Iterative Selection of Gaussian Mixture Priors
arXiv:2603.10935v2 Announce Type: replace-cross Abstract: Variational autoencoders (VAEs) frequently suffer from posterior collapse, where latent variables become uninformative and the…
Bielik-Minitron-7B: Compressing Large Language Models via Structured Pruning and Knowledge Distillation for the Polish Language
arXiv:2603.11881v1 Announce Type: cross Abstract: This report details the creation of Bielik-Minitron-7B, a compressed 7.35B parameter version of the Bielik-11B-v3.0…
The Mirror Design Pattern: Strict Data Geometry over Model Scale for Prompt Injection Detection
arXiv:2603.11875v1 Announce Type: cross Abstract: Prompt injection defenses are often framed as semantic understanding problems and delegated to increasingly large…
Towards Robust Speech Deepfake Detection via Human-Inspired Reasoning
arXiv:2603.10725v2 Announce Type: replace-cross Abstract: The modern generative audio models can be used by an adversary in an unlawful manner,…
Agentic Control Center for Data Product Optimization
arXiv:2603.10133v1 Announce Type: new Abstract: Data products enable end users to gain greater insights about their data by providing supporting…
MA-EgoQA: Question Answering over Egocentric Videos from Multiple Embodied Agents
arXiv:2603.09827v2 Announce Type: replace-cross Abstract: As embodied models become powerful, humans will collaborate with multiple embodied AI agents at their…
EvoSchema: Towards Text-to-SQL Robustness Against Schema Evolution
arXiv:2603.10697v1 Announce Type: cross Abstract: Neural text-to-SQL models, which translate natural language questions (NLQs) into SQL queries given a database…
RandMark: On Random Watermarking of Visual Foundation Models
arXiv:2603.10695v1 Announce Type: cross Abstract: Being trained on large and diverse datasets, visual foundation models (VFMs) can be fine-tuned to…