# Empty Shelves or Lost Keys? Recall Is the Bottleneck for Parametric Factuality

arXiv: https://arxiv.org/abs/2602.14080 · HTML: https://mlsanity.com/paper/2602.14080/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 15 Feb 2026.

## Built on by (most cited first, 12 of 12)

- [Thinking to Recall: How Reasoning Unlocks Parametric Knowledge in LLMs](https://mlsanity.com/paper/2603.09906/) (2026)
- [Why Fine-Tuning Encourages Hallucinations and How to Fix It](https://mlsanity.com/paper/2604.15574/) (2026)
- [Hallucination as Commitment Failure: Larger LLMs Misfire Despite Knowing the Answer](https://mlsanity.com/paper/2605.22007/) (2026)
- [Synthetic Mixed Training: Scaling Parametric Knowledge Acquisition Beyond RAG](https://mlsanity.com/paper/2603.23562/) (2026)
- [MoRFI: Monotonic Sparse Autoencoder Feature Identification](https://mlsanity.com/paper/2604.26866/) (2026)
- [Language models struggle with compartmentalization](https://mlsanity.com/paper/2605.19284/) (2026)
- [LMs as Task-Specific Knowledge Bases: An Interpretability Analysis](https://mlsanity.com/paper/2606.27237/) (2026)
- [From Storage to Access: Verifiable Activation of Parametric Knowledge in LLMs via Explicit Priming and Implicit Reasoning](https://mlsanity.com/paper/2608.18581/) (2026)
- [Knowing but Not Saying: Preventing Factual Access Failures in LLM SFT via Recall-Anchored Distillation](https://mlsanity.com/paper/2608.20794/) (2026)
- [Stick to What You Know: A Study of Knowledge-Aligned Supervised Fine-Tuning](https://mlsanity.com/paper/2608.30987/) (2026)
- [Why Pretraining Fails to Share Cross-Lingual Knowledge](https://mlsanity.com/paper/2609.19291/) (2026)
- [Knowing When Thinking Is Not Enough: Teaching Small Reasoning Models to Reason Beyond Their Parametric Knowledge](https://mlsanity.com/paper/2609.34327/) (2026)

## Abstract (arXiv)

Standard factuality evaluations of LLMs treat all errors alike, obscuring whether failures arise from missing knowledge (empty shelves) or from limited access to encoded facts (lost keys). We propose a behavioral framework that profiles factual knowledge at the level of facts rather than questions, characterizing each fact by whether it is encoded, and then by how accessible it is: cannot be recalled, can be directly recalled, or can only be recalled with inference-time computation (thinking). To support such profiling, we introduce WikiProfile, a new benchmark constructed via an automated pipeline with a prompted LLM grounded in web search. Across 4 million responses from 13 LLMs, we find that encoding is nearly saturated in frontier models on our benchmark, with GPT-5 and Gemini-3 encoding 95--98% of facts. However, recall remains a major bottleneck: many errors previously attributed to missing knowledge instead stem from failures to access it. These failures are systematic and disproportionately affect long-tail facts and reverse questions. Finally, we show that thinking improves recall and can recover a substantial fraction of failures, indicating that future gains may rely less on scaling and more on methods that improve how models utilize what they already encode.