AIModels.fyi

Can doctors trust AI diagnostic tools enough to delegate tasks?
Towards physician-centered oversight of conversational diagnostic AI
Can seeing the document like a human dramatically boost a RAG system's IQ?
Vision-Guided Chunking Is All You Need: Enhancing RAG with Multimodal Document Understanding
Can AI reconstruct super-slow-motion 4D models from regular speed multi-camera video?
4DSloMo: 4D Reconstruction for High Speed Scene with Asynchronous Capture
An embarrassingly simple defense against LLM abliteration attacks
Defending AI systems against a new form of attack
Questioning the role of "chains of thought"
Beyond semantics: The unreasonable effectiveness of reasonless intermediate tokens
Zero-shot voice cloning without transcription
MiniMax-Speech: Intrinsic zero-shot text-to-speech with a learnable speaker encoder
X-Transfer Attacks
Towards super transferable adversarial attacks on CLIP
Big companies are hacking AI's top leaderboard
Goodhart's Law comes for LLMs
Using AI agents to make more realistic 3D scenes
Scenethesis is an agentic framework for 3D scene generation
Recursively summarizing enables long-term dialog memory in LLMs
The challenge of long-term memory in AI conversations