AI safety and alignment

Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.

10,801 stories · RSS feed

Hugging Face Trending Papers
Jun 11

TetherCache: Stabilizing Autoregressive Long-Form Video Generation with Gated Recall and Trusted Alignment

Autoregressive video diffusion models provide a natural formulation for streaming and variable-length video generation by conditioning newly generated frames on previously generated content. However, extending these models to minute-level generation remains challenging: the limited KV-cache budget prevents the model from retaining the full history, while repeatedly conditioning on self-generated frames induces a context distribution shift that accumulates over time, leading to visual artifacts, quality degradation, and temporal drift.

Hugging Face Trending Papers
Jun 11

MARS: Margin-Adversarial Risk-controlled Stopping for Parallel LLM Test-time Scaling

Parallel test-time scaling samples many reasoning traces and majority-votes their answers, improving LLM accuracy but requiring traces to run to completion, incurring substantial computational overhead. We observe that probing partial traces at intermediate checkpoints can extract current answers without disrupting generation, revealing an evolving aggregate vote.

Hugging Face Trending Papers
Jun 11

Bridging Modal Isolation in Interleaved Thinking: Supervising Modality Transitions via Stepwise Reinforcement

Interleaved thinking, where a unified multimodal model alternates between textual reasoning and visual generation, has shown promise on spatial and physical tasks. However, in complex long-chain scenarios, we identify a fundamental failure mode: generated images diverge from the textual context while subsequent text ignores the visual evidence, causing the two modalities to alternate without genuinely informing each other.

arXiv Machine Learning
Jun 11

Urban Heat MiniCubes: An AI-Ready dataset for urban heat research

arXiv:2606. 11534v1 Announce Type: cross Abstract: Urban heat is amplified by impermeable surfaces and heterogeneous built environments, yet street-level variability remains difficult to quantify because multi-sensor observations are rarely available in consistent, analysis-ready form at the necessary spatiotemporal scales.

By Jonathan Starfeldt, Maria J. Molina, Alexander Kerr, Adam Yang, Thomas R. H. Holmes, Christopher R. Hain
arXiv Machine Learning
Jun 11

GraspLLM: Towards Zero-Shot Generalization on Text-Attributed Graphs with LLMs

arXiv:2606. 11898v1 Announce Type: cross Abstract: Research on Text-Attributed Graphs (TAGs) has gained significant attention recently due to its broad applications across various real-world data scenarios, such as citation networks, e-commerce platforms, social media, and web pages.

By Hengyi Feng, Zeang Sheng, Meiyi Qiang, Meiyi Qiang, Wentao Zhang
arXiv AI
Jun 11

Privacy-Preserving Federated Autoencoder for ECG Anomaly Detection on Edge Devices

arXiv:2606. 11556v1 Announce Type: cross Abstract: Continuous electrocardiography (ECG) monitoring could surface rhythm abnormalities before they escalate into cardiovascular events.

By Kaan Arda Akyol, Jakub Kacper Szel\k{a}g, Aydin Abadi, Maha Alghamdi, Ghadah Albalawi, Ghouse Ibrahim Kaleelullah, Hilal Tutus, Sarah Al Subaiei, Shardul Kapse, Syed Mohammed Raheeb, Mujeeb Ahmed, Rehmat Ullah
arXiv AI
Jun 11

EKF-Based Depth Camera and Deep Learning Fusion for UAV-Person Distance Estimation and Following in SAR Operations

arXiv:2602. 20958v2 Announce Type: replace-cross Abstract: Vision-based Unmanned Aerial Vehicles (UAVs) frameworks aid human search tasks by detecting and recognizing specific individuals, then tracking and following them while maintaining a safe distance.

By Luka \v{S}iktar, Branimir \'Caran, Bojan \v{S}ekoranja, Marko \v{S}vaco