Tag: AI Benchmarks
Browse our exclusive articles!
SciIntegrity-Bench: Benchmarking Academic Integrity in AI Research
Discover SciIntegrity-Bench, a benchmark evaluating academic integrity challenges in AI scientist systems and improving ethical AI research practices.
FormalRewardBench: Benchmark for Theorem Proving Rewards
Discover FormalRewardBench, a new benchmark to evaluate reward models in formal theorem proving, enhancing AI's proof evaluation and learning.
EnactToM: Benchmarking Functional Theory of Mind in AI Agents
Discover EnactToM, a 300-task benchmark evaluating AI agents' functional Theory of Mind in 3D environments with dynamic difficulty and formal verification.
Ambig-DS: Benchmarking Task Ambiguity in Data Science AI
Discover Ambig-DS, a new benchmark evaluating task-framing ambiguity in data-science agents to improve AI accuracy and reliability.
SeePhys Pro: Benchmarking Multimodal RLVR in Physics Reasoning
Explore SeePhys Pro, a new benchmark analyzing modality transfer and blind training effects in multimodal reinforcement learning for physics reasoning.