Self-enhancement bias
Self-enhancement bias is an evaluator’s tendency to favor outputs produced by itself or similar systems, regardless of their relative quality.
LLM Evaluation and Alignment, The Foundational Ideas
Self-enhancement bias is a tendency for a language-model judge to favor outputs generated by itself or by models from the same model family, including when those outputs are not objectively better.
- How it is used
- Diagnoses a failure mode in LLM-as-a-judge evaluation before trusting model-generated preferences.
- How deep it goes
- Explains definition, comparative evidence, stylistic mechanism, caveats, and ensemble mitigation, but does not provide a measurement procedure.
- Examples
- Uses GPT-4, Claude-v1, and GPT-3.5 win-rate comparisons plus an RLHF feedback-loop scenario.
- What you can do after
- Diagnose self-preference and select diverse judge ensembles instead of treating one model’s judgment as ground truth.
Read how LLM Evaluation and Alignment, The Foundational Ideas explains self-enhancement bias →
AI Model Evaluation
Self-preference bias is an evaluator's tendency to favor outputs that resemble its own characteristic style, structure, tone, or reasoning patterns.
- How it is used
- Guides LLM-as-a-judge evaluation when stylistic alignment could distort comparisons between outputs.
- How deep it goes
- Defines the bias, identifies style-quality confusion, and gives rubric and debate-based mitigations without detailing implementation beyond call structure.
- Examples
- Uses contrasting concise and verbose responses plus single-call and multi-call argue-then-judge evaluations.
- What you can do after
- Apply it via specific evaluation rubrics and debate-style judging that separate advocacy from final comparison.
Read how AI Model Evaluation explains self-enhancement bias →
Also covered in
- Machines that ThinkIn early access (MEAP)
- Retrieval Augmented Generation, The Foundational IdeasIn early access (MEAP)
- Reinforcement Learning from Human Feedback
What to know first
- G-EvalG-Eval is an LLM-based technique for evaluating generated text against natural-language criteria, often by generating evaluation steps and assigning scores.
- human-in-the-loopHuman-in-the-loop is a system or workflow design in which people participate in reviewing, guiding, correcting, or authorizing an automated process.
- large language modelA large language model is a machine-learning model trained on extensive text data to process and generate natural language.
Related concepts
- LLM-as-a-judgeuses · LLM-as-a-judge is an evaluation technique in which a language model assesses or compares outputs according to specified criteria.
- biaskind of · Bias is a systematic tendency or deviation that distorts measurements, estimates, judgments, or outcomes relative to a relevant reference or target.
- rubricaddressed by · A rubric is a structured set of explicit criteria and guidance for judging the quality of an output, artifact, or behavior.
Questions about self-enhancement bias
What is self-enhancement bias?
Self-enhancement bias is an evaluator’s tendency to favor outputs produced by itself or similar systems, regardless of their relative quality.
Which Manning books explain self-enhancement bias?
LLM Evaluation and Alignment, The Foundational Ideas and AI Model Evaluation.
What should I know before learning self-enhancement bias?
Manning books that teach self-enhancement bias usually introduce Claude-v1, G-Eval, human-in-the-loop and large language model first.
What is self-enhancement bias a kind of?
Self-enhancement bias is a kind of bias. Bias is a systematic tendency or deviation that distorts measurements, estimates, judgments, or outcomes relative to a relevant reference or target.
What is self-enhancement bias also called?
Self-enhancement bias is also called self-preference bias.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.

