Here's your daily roundup of the most relevant AI and ML news for January 06, 2026. We're also covering 8 research developments. Click through to read the full articles from our curated sources.
Research & Papers
1. Crafting Adversarial Inputs for Large Vision-Language Models Using Black-Box Optimization
arXiv:2601.01747v1 Announce Type: cross Abstract: Recent advancements in Large Vision-Language Models (LVLMs) have shown groundbreaking capabilities across diverse multimodal tasks. However, these models remain vulnerable to adversarial jailbreak attacks, where adversaries craft subtle perturbat...
Source: arXiv - Machine Learning | 1 hours ago
2. Explainability-Guided Defense: Attribution-Aware Model Refinement Against Adversarial Data Attacks
arXiv:2601.00968v1 Announce Type: new Abstract: The growing reliance on deep learning models in safety-critical domains such as healthcare and autonomous navigation underscores the need for defenses that are both robust to adversarial perturbations and transparent in their decision-making. In th...
Source: arXiv - Machine Learning | 1 hours ago
3. Learning with Monotone Adversarial Corruptions
arXiv:2601.02193v1 Announce Type: new Abstract: We study the extent to which standard machine learning algorithms rely on exchangeability and independence of data by introducing a monotone adversarial corruption model. In this model, an adversary, upon looking at a "clean" i.i.d. dataset, insert...
Source: arXiv - Machine Learning | 1 hours ago
4. MORE: Multi-Objective Adversarial Attacks on Speech Recognition
arXiv:2601.01852v1 Announce Type: cross Abstract: The emergence of large-scale automatic speech recognition (ASR) models such as Whisper has greatly expanded their adoption across diverse real-world applications. Ensuring robustness against even minor input perturbations is therefore critical fo...
Source: arXiv - Machine Learning | 1 hours ago
5. CEE: An Inference-Time Jailbreak Defense for Embodied Intelligence via Subspace Concept Rotation
arXiv:2504.13201v3 Announce Type: replace-cross Abstract: Large language models (LLMs) are widely used for task understanding and action planning in embodied intelligence (EI) systems, but their adoption substantially increases vulnerability to jailbreak attacks. While recent work explores infer...
Source: arXiv - Machine Learning | 1 hours ago
6. RefSR-Adv: Adversarial Attack on Reference-based Image Super-Resolution Models
arXiv:2601.01202v1 Announce Type: cross Abstract: Single Image Super-Resolution (SISR) aims to recover high-resolution images from low-resolution inputs. Unlike SISR, Reference-based Super-Resolution (RefSR) leverages an additional high-resolution reference image to facilitate the recovery of hi...
Source: arXiv - AI | 1 hours ago
7. Game of Coding: Coding Theory in the Presence of Rational Adversaries, Motivated by Decentralized Machine Learning
arXiv:2601.02313v1 Announce Type: new Abstract: Coding theory plays a crucial role in enabling reliable communication, storage, and computation. Classical approaches assume a worst-case adversarial model and ensure error correction and data recovery only when the number of honest nodes exceeds t...
Source: arXiv - Machine Learning | 1 hours ago
8. Tuning without Peeking: Provable Generalization Bounds and Robust LLM Post-Training
arXiv:2507.01752v3 Announce Type: replace Abstract: Gradient-based optimization is the workhorse of deep learning, offering efficient and scalable training via backpropagation. However, exposing gradients during training can leak sensitive information about the underlying data, raising privacy a...
Source: arXiv - Machine Learning | 1 hours ago
About This Digest
This digest is automatically curated from leading AI and tech news sources, filtered for relevance to AI security and the ML ecosystem. Stories are scored and ranked based on their relevance to model security, supply chain safety, and the broader AI landscape.
Want to see how your favorite models score on security? Check our model dashboard for trust scores on the top 500 HuggingFace models.