Injecting New Knowledge into Large Language Models via Supervised Fine-Tuning · HackerPapers discussions

Injecting New Knowledge into Large Language Models via Supervised Fine-Tuning — mentioned by ijk

Sun, 05 Oct 2025 01:58:29 GMT

Adding knowledge works, depending on how to define knowledge and works; given sufficient data you can teach an LLM new things [1]. However, the frontier models keep improving at a quick enough rate that it's often more effective just to wait for the general solution to catch up with your task then to spend months training a model yourself. Unless you need a particular tightly controlled behavior or need a smaller faster model or what have you. Training new knowledge in can get weird [2]. And in-context learning takes literal seconds-to-minutes of time if your information fits in the context…

Paper and discussion history. A mention is not an endorsement.

Injecting New Knowledge into Large Language Models via Supervised Fine-Tuning — mentioned by ijk

Sun, 05 Oct 2025 01:45:15 GMT

That's consistent with other research I've seen, where varied presentation of the data is key to effective knowledge injection [1]. My assumption, based on the research is that training on different prompts but the same answer gives you more robust Q&A behavior; training on variations of how to express the same concept generalizes. Training on the same prompt and different answers gives you creative diversity [2]. [1] https://arxiv.org/abs/2404.00213 [2] https://arxiv.org/abs/2503.17126

Paper and discussion history. A mention is not an endorsement.

Injecting New Knowledge into Large Language Models via Supervised Fine-Tuning — mentioned by ijk

Sun, 01 Jun 2025 18:11:48 GMT

Depends on the definition of "knowledge"; there's a lot of factors that go into it. Some of the common approaches are continued/continual pretraining and model editing (https://arxiv.org/pdf/2502.12598). * Models are bad at learning that A=B implies B=A, let alone more complicated relations; augmenting the dataset with multiple examples with different phrasing/perspectives is important (https://arxiv.org/abs/2404.00213). The frequency that a relation occurs in the dataset affects the results (https://arxiv.org/html/2504.09597v2). * You have to be able to balance preserving existing knowledge…

Paper and discussion history. A mention is not an endorsement.