Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv:2510. 00054v3 Announce Type: replace-cross Abstract: Multimodal Large Language Models (MLLMs) have made significant strides in visual understanding tasks.
arXiv:2606. 05233v1 Announce Type: cross Abstract: Recent computer-using-agent (CUA) red-teaming papers report prompt-injection attack success rates (ASR) of 42-98%, but these headline numbers cluster on retired models and on the most-vulnerable model in each paper's panel.
arXiv:2606. 06109v1 Announce Type: cross Abstract: Entity alignment (EA) aims to identify equivalent entities across heterogeneous knowledge graphs (KGs) and is a key component of knowledge fusion and cross-KG reasoning.
arXiv:2606. 05290v1 Announce Type: cross Abstract: Recent progress in generative modeling has made safety control a central challenge, yet existing approaches remain largely model-specific, requiring retraining or tailored interventions for each new architecture.
arXiv:2606. 06219v1 Announce Type: cross Abstract: End-to-end autonomous driving models often struggle to balance multi-modal maneuver generation with real-time inference constraints.
arXiv:2606. 05177v1 Announce Type: cross Abstract: Existing multimodal safety benchmarks focus solely on visual inputs and cannot assess Omni Large Language Models (LLMs) that process vision, audio, and text.
arXiv:2503. 14295v3 Announce Type: replace-cross Abstract: Recent advancements in audio-driven talking face generation have made great progress in lip synchronization.
arXiv:2606. 05748v1 Announce Type: cross Abstract: Global-scale video moderation faces a dual challenge: the need for fine-grained multi-modal reasoning and the demand for interpretable outputs to support downstream enforcement.
arXiv:2606. 05740v1 Announce Type: new Abstract: Deep neural networks trained under severe class imbalance often exhibit degraded performance, typically attributed to statistical bias.
arXiv:2606. 05966v1 Announce Type: cross Abstract: Understanding and reasoning about the physical world is the foundation of intelligent behavior, yet state-of-the-art vision-language models (VLMs) still fail at causal physical reasoning, often producing plausible but incorrect answers.
arXiv:2606. 06379v1 Announce Type: cross Abstract: Medical vision-language models (VLMs) have shown increasing potential for clinical image interpretation, including lesion detection and report generation.
arXiv:2606. 05408v1 Announce Type: new Abstract: When an LLM repeatedly mutates a program, does it explore new forms or circle back to the same ones?
arXiv:2605. 27887v2 Announce Type: replace Abstract: Large language models (LLMs) have shown strong performance across diverse financial tasks, yet portfolio management (PM), a critical financial decision-making task, remains poorly benchmarked.
arXiv:2606. 06423v1 Announce Type: cross Abstract: Safety-critical traffic scenario generation is essential for evaluating autonomous driving systems under rare but high-risk interactions.
arXiv:2606. 05647v1 Announce Type: new Abstract: AI coding agents are increasingly embedded in real-world software development, collaborating with human developers while gaining broader access to codebases and tools.
arXiv:2606. 05702v1 Announce Type: new Abstract: Recent advancements in Vision-Language Models (VLMs) have significantly enhanced their ability to interpret complex visual semantics, yet their capacity for chronological reasoning remains under-explored.
arXiv:2606. 05805v1 Announce Type: new Abstract: LLM-based guardrails typically safeguard agents by evaluating proposed actions or inputs before execution, producing safety signals such as binary allow/deny decisions, risk categories, and/or explanatory rationales about potential policy violations.
arXiv:2606. 06154v1 Announce Type: new Abstract: Federated fine-tuning of foundation models using Low-Rank Adaptation (LoRA) offers a communication efficient solution for distributed learning.
arXiv:2606. 06396v1 Announce Type: new Abstract: Autonomous driving technology has the potential to reduce the large number of road traffic accidents caused by human error each year, but it also brings new types of risks that need to be evaluated from the aspects of technology, ethics and regulations.
arXiv:2606. 05183v1 Announce Type: cross Abstract: Large language models are increasingly deployed as high-stakes advisors, yet standard alignment benchmarks treat sycophancy as a binary failure mode.