Where do recommendation biases come from in language models?
Do LLM-based recommenders inherit systematic biases from pretraining that differ fundamentally from traditional collaborative filtering systems? Understanding these sources matters for building fairer, more accurate recommendations.
The Wu et al. survey identifies three biases that LLM-based recommendation systems exhibit but traditional recommenders don't. These biases are inherited from the underlying language model and propagate into recommendation behavior regardless of how the LLM is integrated.
Position bias: when item candidates are presented as a textual sequence in the prompt, the LLM systematically prefers items appearing earlier in the order, regardless of actual relevance. The bias comes from the language modeling objective — early tokens have stronger influence on what the model attends to. The same items in different orderings produce different recommendations.
Popularity bias: the LLM has seen popular items mentioned more frequently in pretraining corpora, so it tends to rank them higher in any recommendation list. This is more pervasive than CF popularity bias because it doesn't come from interaction data — it comes from the world's text. Items famous in news, social media, or product reviews get over-recommended whether they're actually relevant or not. Mitigation is hard because addressing the issue requires changing the pretraining corpus, which is upstream of the recommendation deployment.
Fairness bias: pretrained language models exhibit fairness issues related to sensitive attributes (gender, race, age) reflecting training data demographics. These biases pass through into recommendations, where the LLM might systematically recommend differently to users it perceives as belonging to certain demographic groups.
The implication is that LLM-based recommendation isn't just a more capable variant of conventional recommendation — it's a different beast with its own failure modes. Mitigating these biases isn't about adapting CF debiasing techniques; it requires LLM-specific approaches like balanced prompting, popularity-aware decoding, and fairness-conditioned generation. The research community is still working out the specifics.
Inquiring lines that read this note 31
This note is a source for these research framings, grouped by the broader line of inquiry each explores. Scan the bold lines of inquiry; follow any specific question forward.
Why do LLM recommenders underperform collaborative filtering despite their capabilities?- Why do LLM recommenders drop 60 percent recall when missing collaborative signals?
- Which LLM recommender paradigm actually performs best empirically?
- How does collaborative filtering integrate into LLM-based recommendation systems?
- How do different LLM integration paradigms affect inheritance of pretraining biases?
- Can prompt design strategies reduce position bias in language model recommendations?
- Why is popularity bias harder to fix in LLM recommenders than in collaborative filtering?
- Do pretraining biases and traditional selection bias compound in production recommenders?
- Why do LLM recommenders underperform item-only collaborative filtering baselines?
- How does pretraining corpus popularity bias affect LLM recommendation behavior?
- Which deployment domains favor LLM recommenders over traditional collaborative approaches?
- How do recommender metrics drive LLM query refinement in closed-loop training?
- Can LLMs recommend items without seeing the product catalog?
- Can dataset-level debiasing methods fix popularity bias inherited from pretraining?
- Why does inductive bias outweigh model capacity in recommender systems?
- Can post-hoc reranking actually fix popularity bias created during model training?
- What causes position-induced selection bias in recommendation training data?
- Why do review corpora contain biases that affect generated comparisons?
- Why do LLMs show gender bias but humans evaluators do not?
- What other evaluation biases exist in LLM judge systems?
- Why do LLMs systematically prefer text from their own family?
- How do LLM judges' built-in biases influence the policies they help align?
Related concepts in this collection 4
This note in its neighbourhood — explore the map, then jump to a related concept in the list below.
Click a node to walk · click center to open · click Open in graph to see this note in the full knowledge graph
-
Where does LLM recommendation bias actually come from?
Do conversational AI systems inherit popularity bias from their training data or from the datasets they're deployed on? Understanding the source matters for knowing how to fix it.
exemplifies: empirical instance of the popularity-bias prong — measurable at 5% vs 2% on ReDIAL
-
Why do language models ignore temporal order in ranking?
When LLMs rank items based on interaction history, do they actually use sequence order or treat it as a set? Understanding this gap matters for building effective LLM-based recommenders.
extends: order-blindness is a fourth pretraining-inherited bias adjacent to position bias
-
How should language models integrate into recommender systems?
When building recommendation systems with LLMs, should you use them as feature encoders, token generators, or direct recommenders? The choice affects efficiency, bias, and compatibility with existing pipelines.
complements: each integration paradigm inherits these biases differently — direct generation worst, input-augmentation least
-
Why do ranking systems need to model selection bias explicitly?
Explores how training data from current rankers creates feedback loops that reinforce past decisions. Understanding this mechanism helps explain why naive approaches fail in production ranking systems.
complements: traditional recommender selection-bias and LLM-pretraining biases compose — feedback loops in production can amplify both
Related papers in this collection 8
Papers most semantically related to this note, ranked by cosine similarity in the embedding space.
- A Survey on Large Language Models for Recommendation
- Exploring the Impact of Large Language Models on Recommender Systems: An Extensive Review
- A Multi-facet Paradigm to Bridge Large Language Model and Recommendation
- Large Language Models are Zero-Shot Rankers for Recommender Systems
- CoLLM: Integrating Collaborative Embeddings into Large Language Models for Recommendation
- Prompting Large Language Models for Recommender Systems: A Comprehensive Framework and Empirical Analysis
- Consistent Explainers or Unreliable Narrators? Understanding LLM-generated Group Recommendations
- Large Language Models as Conversational Movie Recommenders: A User Study
Original note title
LLM-based recommendation faces three biases inherited from language model pretraining — position popularity and fairness