Repairing Language Model Pipelines by Meta Self-Refining Competing Constraints at Runtime

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Eshghie, Mojtaba
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908449385742336
author Eshghie, Mojtaba
author_facet Eshghie, Mojtaba
contents Language Model (LM) pipelines can dynamically refine their outputs against programmatic constraints. However, their effectiveness collapses when faced with competing soft constraints, leading to inefficient backtracking loops where satisfying one constraint violates another. We introduce Meta Self-Refining, a framework that equips LM pipelines with a meta-corrective layer to repair these competitions at runtime/inference-time. Our approach monitors the pipeline's execution history to detect oscillatory failures. Upon detection, it invokes a meta-repairer LM that analyzes the holistic state of the backtracking attempts and synthesizes a strategic instruction to balance the competing requirements. This self-repair instruction guides the original LM out of a failing refining loop towards a successful output. Our results show Meta Self-Refining can successfully repair these loops, leading to more efficient LM programs.
format Preprint
id arxiv_https___arxiv_org_abs_2507_10590
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Repairing Language Model Pipelines by Meta Self-Refining Competing Constraints at Runtime
Eshghie, Mojtaba
Software Engineering
Artificial Intelligence
Information Retrieval
Language Model (LM) pipelines can dynamically refine their outputs against programmatic constraints. However, their effectiveness collapses when faced with competing soft constraints, leading to inefficient backtracking loops where satisfying one constraint violates another. We introduce Meta Self-Refining, a framework that equips LM pipelines with a meta-corrective layer to repair these competitions at runtime/inference-time. Our approach monitors the pipeline's execution history to detect oscillatory failures. Upon detection, it invokes a meta-repairer LM that analyzes the holistic state of the backtracking attempts and synthesizes a strategic instruction to balance the competing requirements. This self-repair instruction guides the original LM out of a failing refining loop towards a successful output. Our results show Meta Self-Refining can successfully repair these loops, leading to more efficient LM programs.
title Repairing Language Model Pipelines by Meta Self-Refining Competing Constraints at Runtime
topic Software Engineering
Artificial Intelligence
Information Retrieval
url https://arxiv.org/abs/2507.10590