Topic
in-context learning
Can In-Context Learning Enable Efficient Data Exploration for Enterprise AI?
A research paper investigates whether in-context learning (ICL) can enable intrinsic curiosity—automated data selection—without costly gradient updates. The authors prove that in general Markov decision processes, ICL-based rewards cannot unbiasedly estimate learning progress, but in non-temporal settings like active learning, they succeed. Controlled experiments validate the theory.
Where to Place the Query? Unveiling and Mitigating Positional Bias in Diffusion LLMs via Decoding Dynamics
Researchers uncover that query placement is a first-order variable in diffusion large language models (dLLMs), affecting output quality as much as example semantics. They propose a training-free adaptive routing strategy, Auto-ICL, and a novel metric Average Confidence to mitigate positional bias without ground-truth labels.
Study Reveals 27 Error Types in LLM Text-to-SQL, Introduces MapleDoctor Repair Framework
Researchers conducted the first comprehensive study of errors in LLM-based text-to-SQL systems using in-context learning. They identified 27 error types across 7 categories and proposed MapleDoctor, a detection and repair framework that outperforms existing solutions by repairing 13.8% more queries with negligible mis-repairs and reducing repair latency by 67.4%.
Agentic Framework Achieves 91% Numerical Equivalence in PyTorch-to-JAX Migration via In-Context Learning
Researchers propose an autonomous system that combines in-context learning (ICL) with oracle-driven self-debugging to translate deep learning models from PyTorch to JAX. The lightweight pipeline achieves 91% numerical equivalence, far outperforming baseline methods (9%) and instruction-plus-self-debugging (27%). Validated on models including SAM, T5, and Code Whisper.