Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv:2609.08275v1 Announce Type: new Abstract: Recent multi-shot audio-video generators can produce increasingly coherent and cinematic outputs, but coherence does not imply the ability to execute e...
arXiv:2609.09250v1 Announce Type: cross Abstract: A verifier for robot policies reads a candidate behavior and returns a score for how well it did, used both to evaluate vision-language-action polici...
The paper proposes a content‑based addressing scheme for long‑context models that replaces the growing token counter in Rotary Position Embedding (RoPE) with unit‑level addresses derived from the content of each unit. By dividing the token stream into units, the method preserves local RoPE behavior while allowing new units to be addressed via learned content maps, avoiding positional mismatches when extending context length. Experiments on character‑level Tiny Shakespeare show that a model trained on 256‑character contexts achieves lower perplexity at 4096 characters using this scheme, and a second diagnostic demonstrates retrieval of multiple serialized facts.
HoneyRoute is an inference‑serving layer that identifies malicious requests and redirects them to a honeypot model, protecting the production model while collecting adversarial interactions for intelligence. It combines a lightweight streaming router, a dual‑implementation honeypot, and an analysis loop that turns captured interactions into attacker fingerprints for retraining. In evaluations on production traces and a seven‑domain attack corpus, HoneyRoute achieves high detection F1 scores with minimal latency overhead, significantly reduces token consumption during flooding attacks, and improves fidelity‑traceability trade‑offs compared to naive bait injection.
The paper introduces MedSegLatDiff, a diffusion-based framework that combines a variational autoencoder (VAE) with a latent diffusion model for medical image segmentation. By compressing images into a low-dimensional latent space, the method reduces noise and speeds up training, while a weighted cross‑entropy loss preserves tiny structures such as small nodules. Evaluated on ISIC‑2018, CVC‑Clinic, and LIDC‑IDRI datasets, MedSegLatDiff achieves state‑of‑the‑art Dice and IoU scores, generates diverse segmentation hypotheses, and produces confidence maps that enhance interpretability and reliability for clinical deployment.
arXiv:2609.08729v1 Announce Type: new Abstract: The transition from single-core to multi-core architectures in safety-critical embedded systems introduces significant challenges due to inter-core int...
arXiv:2609.05540v1 Announce Type: cross Abstract: Many medical conditions require diagnosis through detailed, multi-context clinical assessment rather than from visual appearance alone. Despite this,...
arXiv:2609.05550v1 Announce Type: cross Abstract: Near-infrared (NIR) video is a promising modality for contactless sleep monitoring, but recent video-based sleep staging methods often use it as a ro...
arXiv:2609.05591v1 Announce Type: cross Abstract: Safety evaluations typically focus on individual agents, but interacting agents can spread harmful information and influence the environment in which...
arXiv:2609.07959v1 Announce Type: cross Abstract: Machine learning models are widely used in clinical applications, social media, law enforcement and critical infrastructure. Verifying whether their...
arXiv:2609.05742v1 Announce Type: cross Abstract: American Sign Language (ASL) generation remains challenging due to limited paired text-ASL motion data and the difficulty of learning motion represen...
arXiv:2609.09486v1 Announce Type: cross Abstract: Fairness auditing of text-to-image diffusion models often requires generating large numbers of images across sampling configurations, making comprehe...
arXiv:2609.05871v1 Announce Type: cross Abstract: Audio-conditioned language models often underuse acoustic cues such as prosody, emotion, and non-speech sounds, raising the question of whether ASR-s...
arXiv:2605.21957v2 Announce Type: replace Abstract: Video anomaly detection is critical for public safety and security, yet remains highly challenging despite extensive research due to large variatio...
arXiv:2609.05899v1 Announce Type: cross Abstract: Aligning large language models with human preferences remains a challenge, primarily due to the critical role of preference data quality in effective...
arXiv:2607.00815v2 Announce Type: replace-cross Abstract: If the certificate produced by a SAT solver is checked by a verified checker, we get a verdict which convinces. But this verdict cannot be na...
arXiv:2609.05575v1 Announce Type: new Abstract: Understanding how concepts are encoded in the internal representations of machine learning models is a central problem in mechanistic interpretability,...
arXiv:2609.06914v1 Announce Type: new Abstract: Large language models (LLMs) have demonstrated strong capabilities across diverse domains, showing considerable potential in medicine. However, their a...
arXiv:2609.06960v1 Announce Type: new Abstract: Invasive neural recordings provide high-fidelity measurements of brain activity, with signals such as intracranial EEG (iEEG) and intracortical spiking...
arXiv:2609.05975v1 Announce Type: cross Abstract: We introduce Chain of Intent, a governance framework for agentic AI at small regulated firms, and validate it against a failure it was built to addre...