Results for "order information"

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

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

Retrieval based on embedding similarity rather than keyword overlap, capturing paraphrases and related concepts.

Constraining outputs to retrieved or provided sources, often with citation, to improve factual reliability.

Techniques to understand model decisions (global or local), important in high-stakes and regulated settings.

Selecting the most informative samples to label (e.g., uncertainty sampling) to reduce labeling cost.

Protecting data during network transfer and while stored; essential for ML pipelines handling sensitive data.

Structured dataset documentation covering collection, composition, recommended uses, biases, and maintenance.

PEFT method injecting trainable low-rank matrices into layers, enabling efficient fine-tuning.

Converts logits to probabilities by exponentiation and normalization; common in classification and LMs.

Attacks that manipulate model instructions (especially via retrieved content) to override system goals or exfiltrate data.

Attacks that infer whether specific records were in training data, or reconstruct sensitive training examples.

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.

Models that process or generate multiple modalities, enabling vision-language tasks, speech, video understanding, etc.

AI focused on interpreting images/video: classification, detection, segmentation, tracking, and 3D understanding.

Bayesian parameter estimation using the mode of the posterior distribution.

Allows gradients to bypass layers, enabling very deep networks.

A narrow hidden layer forcing compact representations.

Prevents attention to future tokens during training/inference.

All possible configurations an agent may encounter.

Extending agents with long-term memory stores.

Models evaluating and improving their own outputs.

Central catalog of deployed and experimental models.

Inferring sensitive features of training data.

Embedding signals to prove model ownership.

Transformer applied to image patches.

GNN using attention to weight neighbor contributions dynamically.

Combining signals from multiple modalities.

Attention between different modalities.