Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv:2507.17448v2 Announce Type: replace-cross Abstract: Retrosynthetic planning is a cornerstone of organic synthesis and drug discovery. Yet existing AI methods often rely on pattern matching rath...
arXiv:2507.04356v3 Announce Type: replace-cross Abstract: Research, innovation and practical capital investment have been increasing rapidly toward the realization of autonomous physical agents. This...
arXiv:2608.23476v1 Announce Type: new Abstract: Narrow fine-tuning on small, domain-specific datasets can produce broad and surprising changes in model behavior-a phenomenon called weird generalizati...
arXiv:2601.23182v2 Announce Type: replace Abstract: Despite the non-autoregressive potential of diffusion language models (dLLMs), existing decoding strategies demonstrate positional bias, failing to...
arXiv:2504.03337v2 Announce Type: replace Abstract: Existing bias mitigation methods for Visual Question Answering (VQA), a typical Artificial intelligence application, endure two main limitations. F...
arXiv:2608.21374v1 Announce Type: new Abstract: Literature reviews are essential to scientific progress, but rigorously evaluating automatically generated reviews remains difficult because many aspec...
arXiv:2602.17493v2 Announce Type: replace-cross Abstract: We develop a method for training neural networks on Boolean data in which the values at all nodes are strictly $\pm 1$, and the resulting mod...
arXiv:2509.23928v3 Announce Type: replace-cross Abstract: Speculative decoding has proven effective for accelerating inference in Large Language Models (LLMs), yet its extension to Vision-Language Mo...
arXiv:2608.23383v1 Announce Type: new Abstract: Video generation is progressing beyond isolated clips toward long-form narratives and interactive worlds, requiring models to preserve identities, foll...
arXiv:2605.17610v2 Announce Type: replace-cross Abstract: The rapid growth of online video platforms and AI-generated content has made reliable video guardrails a key challenge for safety and real-wo...
arXiv:2603.02229v2 Announce Type: replace-cross Abstract: Safety post-training has been studied extensively in single-step "chat" settings where safety typically refers to refusing harmful requests....
arXiv:2608.17795v2 Announce Type: replace Abstract: Text-to-SQL systems are commonly evaluated using ground-truth SQL queries or reference execution results, but such supervision is unavailable at in...
arXiv:2608.22965v1 Announce Type: new Abstract: Accurate extrinsic calibration between event-based and frame-based cameras remains a practical bottleneck for heterogeneous stereo systems. Existing ap...
The paper evaluates using large language model (LLM) juries to review code generated from natural language queries, focusing on MySQL text-to-SQL tasks. It benchmarks 15 open models, selects the top six, and constructs unanimous committees of varying sizes to accept a query only when all members agree. The study finds that single-model judges are inconsistent, while small unanimous committees of strong models can reduce false accepts without discarding many correct queries, and that committee composition significantly influences performance.
The paper introduces Scaling Electronic Health Record Foundation Models for Population Health Management, a large‑scale model trained on billions of medical events from over 5 million patients in Taiwan and the United States. By aligning ICD codes across different health systems, the model achieves strong scaling and generalization across 11 chronic disease prediction tasks, outperforming tree‑based, general, and biomedical language models with high sensitivity at 99% specificity. It also demonstrates superior few‑shot performance on the EHRShot benchmark and shows that cross‑system alignment provides a stronger pretraining signal than single‑site duplication in data‑limited scenarios.
arXiv:2608.22411v1 Announce Type: new Abstract: Social interaction increasingly takes place in multicultural settings, where individuals may draw on multiple cultural influences and adapt their commu...
arXiv:2602.07008v5 Announce Type: replace Abstract: Reliable models should not only predict correctly, but also base their decisions on acceptable evidence. However, conventional supervised learning...
LangDriveCTRL is a natural‑language‑controllable framework that edits real‑world driving videos by representing each video as an explicit 3D scene graph, separating a static background from dynamic object nodes. It employs a feedback‑driven agentic pipeline where an Orchestrator translates user instructions into executable graphs that coordinate specialized multi‑modal agents—Object Grounding, Behavior Editing, and Behavior Reviewer—to align text with scene nodes, generate and refine multi‑object trajectories, and ensure photorealism through a video diffusion tool and Video Reviewer. The system supports object node editing (removal, insertion, replacement) and multi‑object behavior editing, achieving nearly twice the instruction alignment of prior state‑of‑the‑art methods while preserving photorealism, structural integrity, and traffic realism.
arXiv:2608.21444v1 Announce Type: new Abstract: Multi-drone systems are increasingly positioned for safety-critical missions such as search and rescue (SAR) and critical infrastructure monitoring. Ye...
arXiv:2604.09305v4 Announce Type: replace Abstract: Traffic accidents are a leading cause of fatalities and injuries across the globe. Therefore, the ability to anticipate hazardous situations in adv...