# ReFormBench: Benchmarking the Stability of Response-Conditioned Format Correction in Large Language Models

arXiv: https://arxiv.org/abs/2610.19693 · HTML: https://mlsanity.com/paper/2610.19693/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 14 of its 24 references that are in the map.
- First posted on arXiv 14 Oct 2026.

## Builds on (most cited first, 14 of 14)

- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [gpt-oss-120b & gpt-oss-20b Model Card](https://mlsanity.com/paper/2508.10925/) (2025)
- [DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence](https://mlsanity.com/paper/2606.19348/) (2026)
- [ELI5: Long Form Question Answering](https://mlsanity.com/paper/1907.09190/) (2019)
- [AI Chains: Transparent and Controllable Human-AI Interaction by Chaining Large Language Model Prompts](https://mlsanity.com/paper/2110.01691/) (2021)
- [WildBench: Benchmarking LLMs with Challenging Tasks from Real Users in the Wild](https://mlsanity.com/paper/2406.04770/) (2024)
- [Grammar-Constrained Decoding for Structured NLP Tasks without Finetuning](https://mlsanity.com/paper/2305.13971/) (2023)
- [Benchmarking Complex Instruction-Following with Multiple Constraints Composition](https://mlsanity.com/paper/2407.03978/) (2024)
- [FOFO: A Benchmark to Evaluate LLMs' Format-Following Capability](https://mlsanity.com/paper/2402.18667/) (2024)
- [Learning to Generate Structured Output with Schema Reinforcement Learning](https://mlsanity.com/paper/2502.18878/) (2025)
- [StructEval: Benchmarking LLMs' Capabilities to Generate Structural Outputs](https://mlsanity.com/paper/2505.20139/) (2025)
- [Verifiable Format Control for Large Language Model Generations](https://mlsanity.com/paper/2502.04498/) (2025)
- [ThoughtTrace: Understanding User Thoughts in Real-World LLM Interactions](https://mlsanity.com/paper/2605.20087/) (2026)
- [Harness-IF: Evaluating Instruction Following Across Instruction Surfaces in Coding Agents](https://mlsanity.com/paper/2608.11727/) (2026)

## Abstract (arXiv)

Reliable use of large language models requires responses to satisfy formatting requirements across changing instructions. Standard compliance scores do not distinguish preserving a model's observed format from correcting it, or reveal whether that correction holds across variants. We introduce ReFormBench, a benchmark that uses each model's unconstrained response to the same question as its formatting baseline. It links targets by feature and require/forbid mode, evaluating their instantiated requirements across variants. ReFormBench comprises 5,000 difficulty-screened three-constraint test instances derived from 910 source questions and covering 66 formatting features. TCPR measures complete instruction compliance, while EMCR measures individual and joint correction of targets opposed to the annotated baseline state. Across ten models, EMCR ranges from 34.04% to 64.72%, and the highest three-target correction estimate is 50.02%. Compliance and correction scores yield different model rankings and task-level profiles. A prompting experiment on Qwen3.5-9B improves both metrics, and the catalog covers over 95% of extracted instances on four additional models. ReFormBench makes response-conditioned correction across instruction variants a measurable dimension of format following. The code and data are available at https://anonymous.4open.science/r/ReFormBench/.