# SELF-[IN]CORRECT: LLMs Struggle with Discriminating Self-Generated Responses

arXiv: https://arxiv.org/abs/2404.04298 · HTML: https://mlsanity.com/paper/2404.04298/

## Key facts

- Cited by 26 papers in the map (data through 14 Oct 2026).
- Placed in Self-rewarding alignment and self-training, one of 13 directions in Synthetic data and self-improvement for LLMs.
- Builds on 41 of its 53 references that are in the map.
- First posted on arXiv 4 Apr 2024.

## Placed in

- Synthetic data and self-improvement for LLMs / [Self-rewarding alignment and self-training](https://mlsanity.com/synth/self-rewarding-alignment-and-self-training/)

## Builds on (most cited first, 30 of 41)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [Self-Refine: Iterative Refinement with Self-Feedback](https://mlsanity.com/paper/2303.17651/) (2023)
- [Scaling Instruction-Finetuned Language Models](https://mlsanity.com/paper/2210.11416/) (2022)
- [TruthfulQA: Measuring How Models Mimic Human Falsehoods](https://mlsanity.com/paper/2109.07958/) (2021)
- [TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension](https://mlsanity.com/paper/1705.03551/) (2017)
- [Self-Instruct: Aligning Language Models with Self-Generated Instructions](https://mlsanity.com/paper/2212.10560/) (2022)
- [Voyager: An Open-Ended Embodied Agent with Large Language Models](https://mlsanity.com/paper/2305.16291/) (2023)
- [Large Language Models Cannot Self-Correct Reasoning Yet](https://mlsanity.com/paper/2310.01798/) (2023)
- [ChatGPT Outperforms Crowd-Workers for Text-Annotation Tasks](https://mlsanity.com/paper/2303.15056/) (2023)
- [Super-NaturalInstructions: Generalization via Declarative Instructions on 1600+ NLP Tasks](https://mlsanity.com/paper/2204.07705/) (2022)
- [CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquing](https://mlsanity.com/paper/2305.11738/) (2023)
- [The Flan Collection: Designing Data and Methods for Effective Instruction Tuning](https://mlsanity.com/paper/2301.13688/) (2023)
- [Can Large Language Models Be an Alternative to Human Evaluations?](https://mlsanity.com/paper/2305.01937/) (2023)
- [Self-Rewarding Language Models](https://mlsanity.com/paper/2401.10020/) (2024)
- [Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Models](https://mlsanity.com/paper/2401.01335/) (2024)
- [Self-critiquing models for assisting human evaluators](https://mlsanity.com/paper/2206.05802/) (2022)
- [SeqGAN: Sequence Generative Adversarial Nets with Policy Gradient](https://mlsanity.com/paper/1609.05473/) (2016)
- [PromptSource: An Integrated Development Environment and Repository for Natural Language Prompts](https://mlsanity.com/paper/2202.01279/) (2022)
- [Generating Sequences by Learning to Self-Correct](https://mlsanity.com/paper/2211.00053/) (2022)
- [Automatically Correcting Large Language Models: Surveying the landscape of diverse self-correction strategies](https://mlsanity.com/paper/2308.03188/) (2023)
- [Embers of Autoregression: Understanding Large Language Models Through the Problem They are Trained to Solve](https://mlsanity.com/paper/2309.13638/) (2023)
- [On the Self-Verification Limitations of Large Language Models on Reasoning and Planning Tasks](https://mlsanity.com/paper/2402.08115/) (2024)
- [When Benchmarks are Targets: Revealing the Sensitivity of Large Language Model Leaderboards](https://mlsanity.com/paper/2402.01781/) (2024)
- [Can Large Language Models Really Improve by Self-critiquing Their Own Plans?](https://mlsanity.com/paper/2310.08118/) (2023)
- [The Generative AI Paradox: "What It Can Create, It May Not Understand"](https://mlsanity.com/paper/2311.00059/) (2023)
- [GPT-4 Doesn't Know It's Wrong: An Analysis of Iterative Prompting for Reasoning Problems](https://mlsanity.com/paper/2310.12397/) (2023)

## Built on by (most cited first, 26 of 26)

- [ScienceAgentBench: Toward Rigorous Assessment of Language Agents for Data-Driven Scientific Discovery](https://mlsanity.com/paper/2410.05080/) (2024)
- [ScoreFlow: Mastering LLM Agent Workflows via Score-based Preference Optimization](https://mlsanity.com/paper/2502.04306/) (2025)
- [SEED: Self-Evolving On-Policy Distillation for Agentic Reinforcement Learning](https://mlsanity.com/paper/2607.14777/) (2026)
- [Large Language Models Can Self-Improve in Long-context Reasoning](https://mlsanity.com/paper/2411.08147/) (2024)
- [Mind the Gap: Examining the Self-Improvement Capabilities of Large Language Models](https://mlsanity.com/paper/2412.02674/) (2024)
- [DeepPresenter: Environment-Grounded Reflection for Agentic Presentation Generation](https://mlsanity.com/paper/2602.22839/) (2026)
- [Can LLMs Correct Themselves? A Benchmark of Self-Correction in LLMs](https://mlsanity.com/paper/2510.16062/) (2025)
- [Skill Self-Play: Pushing the Frontier of LLM Capability with Co-Evolving Skills](https://mlsanity.com/paper/2607.22529/) (2026)
- [RATIONALYST: Mining Implicit Rationales for Process Supervision of Reasoning](https://mlsanity.com/paper/2410.01044/) (2024)
- [Beyond Introspection: Reinforcing Thinking via Externalist Behavioral Feedback](https://mlsanity.com/paper/2501.01457/) (2024)
- [RankAlign: A Ranking View of the Generator-Validator Gap in Large Language Models](https://mlsanity.com/paper/2504.11381/) (2025)
- [Memory-Augmented Agent Training for Business Document Understanding](https://mlsanity.com/paper/2412.15274/) (2024)
- [Judging Is Not Enumerating: Silent Omissions in LLM-Authored Acceptable Sets](https://mlsanity.com/paper/2608.01000/) (2026)
- [GuessBench: Sensemaking Multimodal Creativity in the Wild](https://mlsanity.com/paper/2506.00814/) (2025)
- [Who Flips? Self- and Cross-Model Counterarguments Reveal Answer Instability in LLMs](https://mlsanity.com/paper/2606.16011/) (2026)
- [Can LLMs Judge Better Than They Generate? Evaluating Task Asymmetry, Mechanistic Interpretability and Transferability for In-Context QA](https://mlsanity.com/paper/2606.28050/) (2026)
- [A Data-Centric Approach for Safe and Secure Large Language Models against Threatening and Toxic Content](https://mlsanity.com/paper/2504.16120/) (2025)
- [Unlocking Recursive Thinking of LLMs: Alignment via Refinement](https://mlsanity.com/paper/2506.06009/) (2025)
- [A Rolling Stone Gathers No Moss: Adaptive Policy Optimization for Stable Self-Evaluation in Large Multimodal Models](https://mlsanity.com/paper/2508.09155/) (2025)
- [Safety Game: Inference-Time Alignment of Black-Box LLMs via Constrained Optimization](https://mlsanity.com/paper/2510.09330/) (2025)
- [Aligning to Illusions: Choice Blindness in Human and AI Feedback](https://mlsanity.com/paper/2603.08412/) (2026)
- [DCGC: Draft-Conditioned Global Correction for Complex Reasoning with Masked Diffusion Models](https://mlsanity.com/paper/2608.25428/) (2026)
- [Reflective Recovery: A Self-Supervised Method for Reasoning by Learning from Mistakes](https://mlsanity.com/paper/2609.19156/) (2026)
- [The Architect, the Adversary, and the Judge: Closed-Loop Generation of Standards-Aligned Assessment Items at Scale](https://mlsanity.com/paper/2609.26087/) (2026)
- [Multi-agent discussion gains less when dissent is withheld](https://mlsanity.com/paper/2609.38324/) (2026)
- [Feedback as an Interface: Controlling Form and Detail in LLM Translation Refinement](https://mlsanity.com/paper/2610.15661/) (2026)

## Abstract (arXiv)

Can LLMs consistently improve their previous outputs for better results? For this to be true, LLMs would need to be better at discriminating among previously-generated alternatives, than generating initial responses. We explore the validity of this hypothesis in practice. We first formulate a unified framework that allows us to compare the generative and discriminative capability of any model on any task. In our resulting experimental analysis of several open-source and industrial LLMs, we observe that models are not reliably better at discriminating among previously-generated alternatives than generating initial responses. This finding challenges the notion that LLMs may be able to enhance their performance only through their own judgment.