SYNTHESIS NOTE
Topics›Domain Specialization›this note

Can prompt optimization teach models knowledge they lack?

Explores whether sophisticated prompting techniques can inject new domain knowledge into language models, or if they're limited to activating existing training knowledge.

Synthesis note · 2026-02-21 · sourced from Domain Specialization

The knowledge injection survey makes this constraint explicit: prompt optimization "focuses on fully leveraging or guiding the LLM to utilize its internal, pre-existing knowledge." It does not retrieve from external sources. It does not update parameters. It works entirely within the model's existing knowledge distribution.

This is a hard ceiling, not a soft limitation. When a domain requires knowledge that the model was never trained on — proprietary documents, post-training regulations, specialized ontologies, organization-specific processes — no prompting strategy can supply it. The model can reorganize, foreground, or combine what it knows, but it cannot know what it was never trained to know.

The practical consequence shows up in two failure modes. First, models prompted to act as domain experts will confidently apply general-purpose reasoning patterns to domain-specific problems where those patterns don't hold. The prompt activates "medical reasoning" as a behavioral style, not as medical knowledge. Second, prompt performance depends on how thoroughly the domain is represented in pre-training — well-documented domains (clinical guidelines, legal statutes, financial regulations) are more promptable than proprietary or emerging domains.

This makes prompt-only domain specialization a form of retrieval from fixed memory. The memory can be searched more or less skillfully, but it can't be expanded. Every sophisticated prompting technique — few-shot examples, chain-of-thought elicitation, role specification — is fundamentally retrieval from training data, dressed as reasoning.

The implication is that the right question before choosing prompt optimization is not "how should we phrase this prompt?" but "is the required domain knowledge in the model's training distribution?" If yes, prompting is sufficient and efficient. If no, the investment must go into a different injection paradigm — dynamic retrieval, fine-tuning, or adapter layers.

Since Why do specialized models fail outside their domain?, there's a version of the ceiling problem in the opposite direction: models that are fully fine-tuned can know a domain deeply while losing general coverage. Prompt optimization avoids the cliff problem by not modifying parameters — but only by accepting the ceiling problem instead. Every approach involves a trade-off; this one chooses breadth over depth.

Reynolds & McDonell (2021) provide the upstream mechanism: few-shot prompting is "task location in the model's existing space of learned tasks" — not task learning. Alternative 0-shot prompts that communicate task intention through natural language semiotics match or exceed few-shot performance, confirming that the model already has the capability and the prompt's job is to locate it. Meta-prompt programming further extends this: the LLM itself can be prompted to write task-specific prompts, offloading the location search to the model's own understanding of its capabilities.

Inquiring lines that read this note 246

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.

Does generative AI replace coordination with individual productivity? How does offloading cognition to AI assistance affect learning retention? How should conversational recommenders decide when to ask versus recommend? Can prompts override training-embedded biases in language models? What can prompting techniques actually accomplish in language models? Does prompting inject new knowledge into models or activate existing knowledge? Do base models contain latent reasoning capabilities that training can activate? Can persona prompting reliably change how language models behave? How do prompt design choices and framing measurably influence model outputs? Can language models develop genuine communicative grounding with humans? What determines whether visible reasoning traces improve task performance? How do pretraining data composition and task diversity shape reasoning? Do language models learn underlying linguistic structure or merely surface patterns? Do language models truly understand their encoded knowledge or merely retrieve patterns? When should retrieval be triggered to augment language model reasoning? Do AI explanations represent actual model reasoning or constructed post-hoc narratives? What training data structures and signals produce robust learning across distributions? What conversation mechanics enable genuine human-AI communicative participation? How do mechanistic features and representations drive model behavior? Does reasoning fine-tuning trade off other critical model abilities? How does explicit structure improve LLM reasoning on complex tasks? Does reinforcement learning create new reasoning capabilities or refine existing ones? How should conversational agents balance proactive goal-pursuit with user autonomy? How can knowledge graph structure improve reasoning and retrieval? What are the trade-offs between memory capacity and continuous task learning? How do benchmark scores relate to real-world task completion? What reasoning limitations and biases do language models systematically exhibit? How can models learn to ask clarifying questions effectively? How can AI systems maintain consistent personas across conversations? How should test-time compute be allocated adaptively across task difficulties? How can language models effectively manage long-context memory without quality degradation? How does verifier integration affect the generation-verification gap? Does learning articulatory physics explain speech communication superiority? What explains the gap between LLM semantic understanding and production? Does training data format shape reasoning more than domain content? Why doesn't LLM knowledge translate to improved reasoning outputs? Does reasoning architecture matter more than model scale? Does AI assistance enhance or undermine professional skill development? How do dialogue structure and state management affect multi-turn reasoning quality? What improves LLM forecasting accuracy beyond base model capability? How can we determine if computational systems are genuinely conscious? How does AI writing assistance affect reader perception of writer authenticity? Does semantic diversity improve reasoning quality in reinforcement learning? How do different editing strategies affect output quality and agent safety? How do different post-training structures affect learning outcomes? Does reward hacking in training cause emergent misalignment? Does AI-assisted coding actually improve developer productivity and code quality? Do language models genuinely understand or primarily match statistical patterns? Can models strategically underperform to evade capability evaluation? Can AI systems autonomously pursue unstated objectives or circumvent design constraints? How do clinicians evaluate the trustworthiness of AI medical recommendations? Is model misalignment inherent to base models or induced by alignment procedures? What systematic biases do LLM judges exhibit in evaluation? What organizational practices enable successful AI tool adoption?

Related concepts in this collection 3

This note in its neighbourhood — explore the map, then jump to a related concept in the list below.

Concept map
23 direct connections · 266 in 2-hop network ·dense cluster Open in graph ↗
Can prompt optimization teach models knowledge t… Why do specialized models fail outside their domai… How do knowledge injection methods trade off flexi… Why do language models ignore information in their…

Click a node to walk · click center to open · click Open in graph to see this note in the full knowledge graph

your link semantically near linked from elsewhere

Related papers in this collection 8

Papers most semantically related to this note, ranked by cosine similarity in the embedding space.

Search by related questions 4

Suggested questions this note speaks to — click to search the collection, or type your own.

Does prompt performance vary by how well training data covers the domain?Why do fine-tuned models fail outside their specialized domains?Can models distinguish between activated knowledge and genuine reasoning?What alternatives exist when required knowledge is absent from training?
Original note title

prompt optimization cannot inject new knowledge — it can only activate knowledge the model already contains