Social Intelligence Benchmark
Summary
This paper introduces a new benchmark for evaluating social intelligence in AI systems, measuring their ability to understand and respond to social cues and interactions.
Similar Articles
ASI-Bench: At the Dawn of Artificial Superintelligence
ASI-Bench is a new benchmark designed to evaluate AI systems' capabilities in innovative exploration and autonomous scientific execution across 11 scientific domains, revealing current AI's heavy dependence on human guidance.
Can Agents Read the Room? Benchmarking Visual Social Intelligence in Multimodal Simulation
This paper introduces AgentViSS, a benchmark evaluating visual social intelligence in multimodal social simulation, containing 240 scenarios with aligned visual-textual evidence. Evaluating seven recent MLLMs reveals a gap between local role enactment and visually grounded interaction management.
SpeechEQ: Benchmarking Emotional Intelligence Quotient in Socially Aware Voice Conversational Models
SpeechEQ introduces a benchmark and dataset for evaluating emotional intelligence in speech-language models, covering 15 EQ subscales across 2,265 dialogues. Experiments reveal current models struggle with paralinguistic cues, exhibiting text-reliant shortcuts and other limitations.
JarvisBench: Always-on Intelligence Between Humans and Agents
JarvisBench introduces a benchmark for evaluating the coordination between humans and AI agents, focusing on attention allocation in long-horizon tasks. It provides a reference implementation with a full-duplex speech interface.
Measuring What Matters: Benchmarking Generative, Multimodal, and Agentic AI in Healthcare
This paper presents a structured framework for benchmarking generative, multimodal, and agentic AI in healthcare, addressing the gap between high benchmark scores and real-world clinical reliability, safety, and relevance.