# User Feedback Provides a Unique Signal that LLMs Can not Detect

arXiv: https://arxiv.org/abs/2609.02859 · HTML: https://mlsanity.com/paper/2609.02859/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 17 of its 30 references that are in the map.
- First posted on arXiv 2 Sep 2026.

## Builds on (most cited first, 17 of 17)

- [G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment](https://mlsanity.com/paper/2303.16634/) (2023)
- [Large Language Models are not Fair Evaluators](https://mlsanity.com/paper/2305.17926/) (2023)
- [Can Large Language Models Be an Alternative to Human Evaluations?](https://mlsanity.com/paper/2305.01937/) (2023)
- [Large Language Models Can Self-Improve](https://mlsanity.com/paper/2210.11610/) (2022)
- [From Crowdsourced Data to High-Quality Benchmarks: Arena-Hard and BenchBuilder Pipeline](https://mlsanity.com/paper/2406.11939/) (2024)
- [JudgeLM: Fine-tuned Large Language Models are Scalable Judges](https://mlsanity.com/paper/2310.17631/) (2023)
- [Learning from Dialogue after Deployment: Feed Yourself, Chatbot!](https://mlsanity.com/paper/1901.05415/) (2019)
- [Self-Adapting Language Models](https://mlsanity.com/paper/2506.10943/) (2025)
- [WildFeedback: Aligning LLMs With In-situ User Interactions And Feedback](https://mlsanity.com/paper/2408.15549/) (2024)
- [Aligning Language Models from User Interactions](https://mlsanity.com/paper/2603.12273/) (2026)
- [Learning to Simulate Natural Language Feedback for Interactive Semantic Parsing](https://mlsanity.com/paper/2305.08195/) (2023)
- [User Feedback in Human-LLM Dialogues: A Lens to Understand Users But Noisy as a Learning Signal](https://mlsanity.com/paper/2507.23158/) (2025)
- [Naturally Occurring Feedback is Common, Extractable and Useful](https://mlsanity.com/paper/2407.10944/) (2024)
- [The Era of Real-World Human Interaction: RL from User Conversations](https://mlsanity.com/paper/2509.25137/) (2025)
- [The Future of Open Human Feedback](https://mlsanity.com/paper/2408.16961/) (2024)
- [The ShareLM Collection and Plugin: Contributing Human-Model Chats for the Benefit of the Community](https://mlsanity.com/paper/2408.08291/) (2024)
- [Mediocrity is the key for LLM as a Judge Anchor Selection](https://mlsanity.com/paper/2603.16848/) (2026)

## Abstract (arXiv)

Harnessing naturally occurring feedback from user interactions offers a promising learning signal for Large Language Models (LLMs). However, recent studies suggest this feedback is inherently noisy and difficult to leverage effectively. We challenge this conception by demonstrating that user feedback is a highly actionable signal for improvement, and that its perceived ineffectiveness stems from a systematic bias in current evaluation paradigms. To isolate the usefulness of feedback, we construct synthetic data with a definitive ground truth, alongside naturalistic data to validate that our findings hold in real-world scenarios. By comparing model revisions generated with and without access to feedback across both settings, we show that feedback-informed revisions resolve targeted issues at significantly higher rates than baseline revisions. Finally, we expose the root of the evaluation bias: when a model successfully fixes an issue exclusively due to feedback, LLM judges frequently fail to identify the genuinely corrected response, systematically preferring inferior baseline outputs instead.