Variable names provide priors: semantic grounding enables data-efficient numerical prediction by large language models
cs.AI, cs.CL, cs.LG
Submitted: 2026-11-02
Updated: 2026-11-02
Terminology
Sources
- GPT-4 Technical Report
- In-Context Function Learning in Large Language Models
- AutoElicit: Using Large Language Models for Expert Prior Elicitation in Predictive Modelling
- Larger language models do in-context learning differently
- In-Context Examples Suppress Scientific Knowledge Recall in LLMs
- Elephants Never Forget: Memorization and Learning of Tabular Data in Large Language Models
- An Explanation of In-context Learning as Implicit Bayesian Inference
- Recasting Gradient-Based Meta-Learning as Hierarchical Bayes
- Transformers Can Do Bayesian Inference
- MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformers
- TabPFN-3: Technical Report
- CARTE: Pretraining and Transfer for Tabular Learning
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents