Here's your daily roundup of the most relevant AI and ML news for October 01, 2026. We're also covering 7 research developments. Click through to read the full articles from our curated sources.
Research & Papers
1. CoER: Defending against Adaptive Indirect Prompt Injection via Adversarial Co-Evolution and Refinement
arXiv:2609.07529v3 Announce Type: replace Abstract: Language-model agents are vulnerable to indirect prompt injection (IPI) during tool use: adversarial instructions hidden in untrusted tool outputs can covertly redirect legitimate task execution. Existing work often trains and evaluates defense...
Source: arXiv - Machine Learning | 10 hours ago
2. Inference-Layer Security: Defending Against Adversarial Inference and Infrastructure Abuse
arXiv:2609.38239v1 Announce Type: cross Abstract: A Technical Report: Operating a large language model (LLM) as a service requires more than inference infrastructure: the provider must also defend against adversarial interactions that seek to exploit the service, including jailbreaking for harmf...
Source: arXiv - Machine Learning | 10 hours ago
3. Kill-Chain Canaries: Stage-Level Tracking of Prompt Injection Across Attack Surfaces and Five Production LLMs
arXiv:2603.28013v4 Announce Type: replace-cross Abstract: Multi-agent LLM systems now read documents, web pages and tool results on behalf of users, yet their resistance to prompt injection is usually reported as one number: did the attack succeed? We introduce a kill-chain canary method that pl...
Source: arXiv - Machine Learning | 10 hours ago
4. The Invisible Language Tax: Token Premiums of French and Regional Languages in 2026 LLM Tokenizers, and a French-Optimized Prototype
arXiv:2609.39001v1 Announce Type: cross Abstract: LLM services are billed per token and context windows are measured in tokens, yet the number of tokens needed for the same content varies across languages. We measure this token premium on seven tokenizers of widely used 2026 models (OpenAI o200k...
Source: arXiv - AI | 10 hours ago
5. Revisiting On-policy Adversarial Black-Box Distillation: Calibrating Groupwise Reward Geometry for Effective Advantage Construction
arXiv:2609.39757v1 Announce Type: new Abstract: Black-box distillation is a practical route for transferring capabilities from API-accessible large language models that expose only text outputs into smaller student models. Recent on-policy adversarial methods such as GAD improve over SeqKD by fo...
Source: arXiv - Machine Learning | 10 hours ago
6. Conformal Adversarial Generative Ensemble
arXiv:2609.38196v1 Announce Type: new Abstract: Accurate time series forecasting is critical across various domains, yet traditional ensemble methods often suffer from the disproportionate influence of extreme forecasts. We introduce the Conformal Adversarial Generative Ensemble (CAGE), a novel ...
Source: arXiv - Machine Learning | 10 hours ago
7. Anchoring Adversarial Trajectories to Data Manifolds: A Bilevel Transfer Optimization Framework
arXiv:2609.38991v1 Announce Type: new Abstract: A key bottleneck in adversarial transfer is a trajectory-level geometric disconnect: ambient gradients often drift away from the intrinsic data manifold, causing surrogate-specific overfitting. To rectify this, we propose Manifold Anchored Bilevel ...
Source: arXiv - Machine Learning | 10 hours ago
Tech & Development
8. Here’s how tech leaders will self-police AI safety under Trump’s deal
We now have the full details of the "morally binding" AI safety deal announced by President Trump yesterday, in which top executives agreed to self-regulate their artificial intelligence technology. The accord, officially titled the Joint Commitment on Frontier Responsibilities, was shared online...
Source: The Verge - AI | 1 day 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.