Results for "reasoning traces"
Stepwise reasoning patterns that can improve multi-step tasks; often handled implicitly or summarized for safety/privacy.
Temporary reasoning space (often hidden).
A broader capability to infer internal system state from telemetry, crucial for AI services and agents.
Interleaving reasoning and tool use.
The field of building systems that perform tasks associated with human intelligence—perception, reasoning, language, planning, and decision-making—via algori...
Maximum number of tokens the model can attend to in one forward pass; constrains long-document reasoning.
Framework for reasoning about cause-effect relationships beyond correlation, often using structural assumptions and experiments.
Simple agent responding directly to inputs.
Agent reasoning about future outcomes.
AI capable of performing most intellectual tasks humans can.
Achieving task performance by providing a small number of examples inside the prompt without weight updates.
Constraining outputs to retrieved or provided sources, often with citation, to improve factual reliability.
Methods for breaking goals into steps; can be classical (A*, STRIPS) or LLM-driven with tool calls.
Capabilities that appear only beyond certain model sizes.
Continuous cycle of observation, reasoning, action, and feedback.
Structured graph encoding facts as entity–relation–entity triples.
What would have happened under different conditions.
System that independently pursues goals over time.
Enables external computation or lookup.
AI systems that perceive and act in the physical world through sensors and actuators.
Software pipeline converting raw sensor data into structured representations.
Understanding objects exist when unseen.
Human-like understanding of physical behavior.
Mathematical guarantees of system behavior.
AI supporting legal research, drafting, and analysis.
Legal right to fair treatment.
Predicting case success probabilities.
AI proposing scientific hypotheses.
A system that perceives state, selects actions, and pursues goals—often combining LLM reasoning with tools and memory.
System-level design for general intelligence.