← Back to Research
arxiv
LLM February 25, 2026 Impact: 8/10

Memorization vs. Reasoning: Updating LLMs with New Knowledge

By Aochong Oliver Li, Tanya Goyal
Large language models (LLMs) encode vast amounts of pre-trained knowledge in their parameters, but updating them as real-world information evolves remains a challenge. Existing methodologies and benchmarks primarily target entity substitutions, failing to capture the full breadth of complex real-world dynamics. In this paper, we introduce Knowledge Update Playground (KUP), an automatic pipeline for simulating realistic knowledge updates reflected in an evidence corpora. KUP's evaluation framework includes direct and indirect probes to both test memorization of updated facts and reasoning over them, for any update learning methods. Next, we present a lightweight method called memory conditioned training (MCT), which conditions tokens in the update corpus on self-generated "memory" tokens during training. Our strategy encourages LLMs to surface and reason over newly memorized knowledge at inference. Our results on two strong LLMs show that (1) KUP benchmark is highly challenging, with the best CPT models achieving less than 2% in indirect probing setting (reasoning) and (2) MCT training significantly outperforms prior continued pre-training (CPT) baselines, improving direct probing (memorization) results by up to 25.4%.
knowledge updatingmemorizationreasoningcontinual learningLLM training
Analysis

Summary

This paper tackles one of the most persistent and underappreciated problems in deployed LLM systems: what happens when the world changes but the model doesn't? Li and Goyal introduce the Knowledge Update Playground (KUP), a benchmark specifically designed to probe whether LLMs can not only memorize newly introduced facts through continued training but actually reason over those facts in novel contexts. The critical distinction the paper draws is between direct probing (can the model recall the updated fact?) and indirect probing (can the model use the updated fact to answer a downstream question it wasn't trained on?). This separation is what makes KUP meaningfully harder than prior benchmarks that essentially tested rote substitution.

Why It Matters

The results are sobering. The best continued pre-training (CPT) models achieve less than 2% accuracy on indirect probing — meaning models can memorize new facts but almost completely fail to deploy them in reasoning chains. This is a devastating finding for any real-world deployment scenario where models need to stay current. Think of a legal assistant that's been updated with new case law, or a medical model updated with new drug interaction data: memorizing the facts is table stakes; reasoning over them is what actually matters. The paper quantifies precisely how wide this gap is, which is valuable groundwork for the field.

Key Contribution: Memory Conditioned Training (MCT)

The proposed remedy, MCT, is elegant in its simplicity. During training on the update corpus, the model is conditioned on self-generated "memory" tokens — essentially forcing the model to explicitly surface relevant memorized knowledge before generating the next token. This is a form of structured self-prompting baked into the training objective rather than added at inference time. The approach improves direct probing by up to 25.4% over CPT baselines, a substantial gain. The implicit hypothesis is that the act of generating the memory token creates stronger associative pathways between the new fact and the contexts in which it becomes relevant.

Implications for the Field

This work has direct implications for retrieval-augmented generation (RAG) vs. parametric knowledge debates, as well as for model editing research. The KUP framework is likely to become a standard evaluation harness for knowledge update methods going forward. More broadly, it reinforces that the gap between memorization and reasoning is not just a capability limitation but an architectural and training challenge that requires deliberate design. For practitioners continually fine-tuning production models, MCT offers a practical, lightweight path toward models that don't just store new knowledge but actually use it.

Key Findings
1
Best CPT models achieve <2% accuracy on indirect (reasoning) probes despite learning new facts
2
MCT improves direct memorization results by up to 25.4% over continued pre-training baselines
3
KUP benchmark separates memorization from reasoning, exposing a critical gap prior benchmarks missed
Impact Score
8/10
Expert Commentary
This is a rigorous, practically relevant paper that exposes a critical and underappreciated failure mode in how we update LLMs — memorizing facts is far easier than reasoning over them, and MCT is a promising lightweight fix worth watching.
D
Dong Tran
AI Researcher. Deep-dive analysis of frontier AI research papers.