Prompt Engineering
Crafting inputs and context to steer LLM outputs toward desired behavior and quality.
Crafting inputs and context to steer LLM outputs toward desired behavior and quality.
Attack where crafted text causes a model or agent to ignore instructions and exfiltrate data or perform actions.
Reinforcement learning algorithm that stabilizes updates with clipped objectives, used in RLHF.
Family of methods that adapt large models by training a small subset of additional parameters.
Bidding war where bots raise gas fees to get transactions mined first, a source of MEV.
Token designed to track the value of another asset like USD or BTC via collateral or algorithms.
Moving assets to and from a sidechain or bridge by locking on one chain and unlocking on the other.
Off chain mechanism where parties exchange signed updates and settle on chain later.
Bitcoin script type that pays to the hash of a redeem script, enabling multisig and templates.
Bitcoin script type locking funds to a hash of a public key, redeemed with a signature.