AI safety and alignment

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

8,629 stories · RSS feed

Hugging Face Trending Papers
Aug 10

Confusion-Geometry Rebalancing for Long-Tailed Adversarial Training

Adversarial training under long tailed distributions suffers from a dual imbalance: the class imbalance skews the training objective toward head classes, and the adversarial inner maximization may further amplify this bias. Existing methods mitigate this issue by correcting class priors or adapting class wise robust supervision, yet they treat each class in isolation and fail to identify which boundaries drive long tailed collapse.

Hugging Face Trending Papers
Aug 10

Warp-free Cross-view Geo-localization via Feature-space Consensus Mining

Cross-view geo-localization is challenging due to drastic viewpoint changes and large appearance discrepancies between street-level and satellite imagery. Although existing methods often use geometric warping to expose co-visible cues, such transformations rely on restrictive spatial assumptions and inevitably introduce severe visual distortions under view-dependent visibility, yielding noisy supervision and fragile correspondences.

Hugging Face Trending Papers
Aug 10

Emotion2Skill: Model-Internal Emotion Signals for Adaptive Skill Selection and Evolution

Skill-based LLM agents select reusable procedures from an external library to solve complex tasks, yet their routing decisions rely entirely on text-level signals such as task descriptions, verbal reflections, and experience-derived rules, while the model's own internal representational state remains unobserved. Recent interpretability work has shown that LLMs maintain linear emotion representations that causally influence behavior; however, these representations have been exploited only for post-hoc analysis or direct output steering, and have not been used to inform agent-level decision-making.

Hugging Face Trending Papers
Aug 10

RL-Native Distillation: Exploiting Scored Trajectories for Few-Step Image Generation

Efficient text-to-image generation requires both reinforcement-learning (RL)-based reward alignment and few-step distillation, yet these procedures are typically performed sequentially, increasing training cost and risking the loss of reward gains during compression. We instead take an RL-native perspective: diffusion RL already generates reward-scored finite-step trajectories, whose intermediate states provide a natural source of distillation supervision rather than a disposable byproduct of sampling.

Hugging Face Trending Papers
Aug 10

FedTVD: Balancing Data Quality and Quantity for Robust Federated Learning

Federated Learning (FL) enables collaborative model training across distributed client devices while preserving data privacy. However, FL faces significant challenges due to data heterogeneity, particularly in terms of label distribution skewness and variations in dataset sizes, which can lead to biased model updates and hinder convergence.

Hugging Face Trending Papers
Aug 10

UNMASK: Discovering and Causally Verifying Spurious Shortcuts in Text Classifiers

Neural language models trained on large crowdsourced corpora frequently exploit spurious surface patterns tied to target labels without true linguistic or causal relevance, boosting benchmark performance while failing on adversarial or out-of-distribution inputs. Existing approaches either require manual specification of the feature vocabulary or automate discovery only partially, leaving the gap between dataset-level correlation and model-level exploitation unaddressed.

Hugging Face Trending Papers
Aug 10

UniMoFlow: Grounding Instruction-Driven 3D Human Motion Editing in Generation

Instruction-driven editing of 3D human motion requires precise spatiotemporal localization, rich semantic grounding, and strict preservation of unmodified content. Existing methods either resort to training-free adaptation of generative models or rely solely on triplet supervision; however, adaptation often yields suboptimal control, and manually curated triplet datasets remain severely limited in scale and semantic diversity.

arXiv AI
Aug 10

Agentic AI: User Empowerment or Enclosure?

arXiv:2608. 06510v1 Announce Type: cross Abstract: Agentic AI promises a more flexible form of digital agency: systems that can act on users' behalf, from filtering content to negotiating prices to selecting services.

By David Gamba, Daniel M. Romero, Grant Schoenebeck
arXiv AI
Aug 10

SoRoMoX: Fast, Differentiable, and Parallelizable Soft Robot Models

arXiv:2608. 06650v1 Announce Type: cross Abstract: Reduced-order models based on Cosserat-rod theory are now well established, and modeling theory is no longer the primary bottleneck in soft-robot control.

By Maximilian St\"olzle, Solange Gribonval, Daniel Feliu-Talegon, Vito Daniele Perfetta, Michele Martini, Chuhan Zhang, Kiwan Wong, Mohammed Tarnini, Anup Teejo Mathew, Federico Renda, Daniela Rus, Cosimo Della Santina
arXiv AI
Aug 10

Artificial Intelligence Can Match Domain Experts in Evidence Extraction and Critical Appraisal of Microbial Oncogenesis Research Publications

arXiv:2608. 07250v1 Announce Type: cross Abstract: Confirmed oncogenic microbes contribute significantly to cancer burden.

By Kaela Kokkas, Hairong Wang, Richard Klein, Nazir A. Ismail, Natalie Irwin, Mohammad Z. Moonsamy, Kubendran Naidoo, Jeremy Nel, Ekene E. Nweke, Raveen Parboosing, Emmanuel K. Sekyi, Rebecca T. van Dorsten, Bruce A. Bassett, Robert F. Breiman