adversarial-learning

This paper studies co-training a monitor alongside an adversarial AI worker to prevent monitor evasion in worker-monitor oversight setups, proving a supervisory characterization via Littlestone dimension and proposing a test-time distillation self-supervision method with code-security stress tests.

This paper resolves an open problem by showing that a single algorithm achieves optimal switching regret for every S against an oblivious adversary in multi-armed bandits.

This paper introduces an adversarial reinforcement learning agent that learns to disrupt self-triggered controllers with sparse Denial-of-Service attacks, outperforming baselines on control systems like Pendulum and Quadrotor2D.

The paper introduces OBJECTION, an inference-time pipeline using adversarial lawyer agents to mitigate guilty bias in legal judgment prediction models, demonstrating a significant reduction in false guilty rates and releasing a new 'Natural Innocent' dataset.

The paper introduces function-symbol grounding in LTN-GANs to handle hard structural constraints in generative models, showing it learns margin distributions more faithfully than existing methods, as validated on high-resolution datasets.

This paper proposes algorithms for sequential prediction under a lying oracle in stochastic and adversarial settings, proving logarithmic regret bounds for the cost function capturing prediction complexity.

This paper proposes learning dynamic classifier-free guidance schedules for diffusion models using adversarial learning to improve text-to-image generation quality by adapting guidance scales to different states.

This paper proposes a linguistic-invariant spoofing detection framework that uses teacher-student adversarial learning and a variational information bottleneck to mitigate linguistic bias, achieving up to a 36.2% relative reduction in equal error rate across nine datasets.

Introduces A3M, a framework combining adaptive deep reinforcement learning, adversarial reasoning, and multi-objective reward design for strategic bidding in repeated auctions, achieving 30-40% regret reduction.

Proposes a Wasserstein-GAN approach for unsupervised calibration of sensor-induced distribution drifts, validated on tracking detector toy models and simulated calorimeter data with aging effects.

This paper introduces Z-Image Turbo++, a two-step image generation model distilled from an eight-step teacher using distribution-aligned adversarial learning, step-decoupled parameterization, and end-to-end training with iterative regularization to narrow the quality gap with multi-step generation.

This paper proposed Distribution-Aligned Adversarial Distillation (DisAAD), a method that uses a lightweight proxy model to estimate uncertainty in black-box LLMs with only 1% of the original model size, achieving reliable quantification without requiring internal parameters or multiple sampling.