BEAKER: An Expert-Curated Benchmark for Embodied Brains in Self-Driving Chemical Laboratories
FrontierPhysics: Evaluating Agents for End-to-End Physics Research
Dull, Dirty, Dangerous? Working Conditions and the Limits of Occupational AI Exposure
Learning to Decode Linear Codes with Generalized Belief Propagation Network
On Turn-Level Best-of-N Sampling for Scaling LLM Agents
CompressionBench: How can your coding agents compress LLMs?
ServerBench: How can your coding agents design specialized LLM serving systems?
Rules, Traces and Routing Statistics: Training and Calibration Data for Open Models on ARC-AGI-2
Task-Efficient Agent Evaluation Using Cached Traces
ConceptARC at Scale: Studying Conceptual Understanding in Reasoning Models
LifeVal: Evaluating AI Model Performance on Personalized Agentic Consumer Tasks
Where Search Agents Go Off the Rails: Graph-Grounded Step-Level Verification
From Certified Pairs to a Certified Hub: Reasoner-Checked Crosswalks for the Information Exchange Standard
Interactive World Simulation with Cosmos for Physical AI
Time-Division Multiplexing: The Operational Rhythm and Temporal Organization Layer of Digital Life
One Scale, Four Texts: Gaps, Overlaps and Silent Repairs in Published Grading Schemes
RAGAM-AI: A Hybrid Reinforcement Learning based Agentic approach powered by Generative AI for Edge Resource Management
Addition by Subtraction: Toward Mitigating Fragile Safety Alignment in Large Language Models
Using Experience Twice: Dynamic Process Rubrics for Privileged Token-Level Credit Assignment
ElecLoad: Adaptive Electricity Load Forecasting Guided by Multiple Events
Describe your interests as you like, be it keywords or sentence, you’ll get relevant papers every day via AI semantic matching.
IJCAI, ICML, CVPR, KDD... Access papers from top AI conferences, all in one tool, for free.
Keep your research organized with our bookmarking feature. Build a well-structured knowledge hub at your fingertips.
Got questions? Ask in ChatDOC with one single click. Use AI to pull data, clarify terms, and verify facts with our precise word-level tracing feature.
