Results for "sim + real"

Performance drop when moving from simulation to reality.

Randomizing simulation parameters to improve real-world transfer.

Combining simulation and real-world data.

Differences between simulated and real physics.

Learning where data arrives sequentially and the model updates continuously, often under changing distributions.

Letting an LLM call external functions/APIs to fetch data, compute, or take actions, improving reliability.

Artificially created data used to train/test models; helpful for privacy and coverage, risky if unrealistic.

Time from request to response; critical for real-time inference and UX.

A dataset + metric suite for comparing models; can be gamed or misaligned with real-world goals.

Low-latency prediction per request.

Two-network setup where generator fools a discriminator.

Control using real-time sensor feedback.

Enables external computation or lookup.

Artificial environment for training/testing agents.

High-fidelity virtual model of a physical system.

Artificial sensor data generated in simulation.

Training with a small labeled dataset plus a larger unlabeled dataset, leveraging assumptions like smoothness/cluster structure.

A mismatch between training and deployment data distributions that can degrade model performance.

A continuous vector encoding of an item (word, image, user) such that semantic similarity corresponds to geometric closeness.

Architecture that retrieves relevant documents (e.g., from a vector DB) and conditions generation on them to reduce hallucinations.

When information from evaluation data improperly influences training, inflating reported performance.

Ensuring model behavior matches human goals, norms, and constraints, including reducing harmful or deceptive outputs.

Model-generated content that is fluent but unsupported by evidence or incorrect; mitigated by grounding and verification.

Automated detection/prevention of disallowed outputs (toxicity, self-harm, illegal instruction, etc.).

Rules and controls around generation (filters, validators, structured outputs) to reduce unsafe or invalid behavior.

Systematic differences in model outcomes across groups; arises from data, labels, and deployment context.

Observing model inputs/outputs, latency, cost, and quality over time to catch regressions and drift.

Constraining model outputs into a schema used to call external APIs/tools safely and deterministically.

Methods for breaking goals into steps; can be classical (A*, STRIPS) or LLM-driven with tool calls.

Attention mechanisms that reduce quadratic complexity.