fix: add reference links for clarification

src/collections/news/2024-04-26-Fine-Tuning-Llama2.md

@@ -20,12 +20,13 @@ grammatical and syntactical rules of Blissymbolics. For example, the English sen
 blue lake" would be represented in Blissymbolics as "present: I move slowly towards lake blue." This structural
 consistency was crucial for the model's learning process.
 
-The fine-tuning was executed on the Cedar cluster, a high-performance computing resource provided by the Digital
-Research Alliance of Canada. By iteratively training the model over four epochs, the team generated multiple
-versions of the fine-tuned model. Each version was tested to evaluate its performance in translating sentences
-between English and Blissymbolics. The evaluation criteria for the converted sentences included semantic coherence,
-novelty, creativity, fluency, and readability. The results were promising, with the models, particularly those
-trained for two and three epochs, demonstrating significant improvements in translation accuracy and fluency.
+The fine-tuning was executed on the Cedar cluster, a high-performance computing resource provided by
+the [Digital Research Alliance of Canada](https://alliancecan.ca). By iteratively training the model over four epochs,
+the team generated multiple versions of the fine-tuned model. Each version was tested to evaluate its performance
+in translating sentences between English and Blissymbolics. The evaluation criteria for the converted sentences
+included semantic coherence, novelty, creativity, fluency, and readability. The results were promising, with the
+models, particularly those trained for two and three epochs, demonstrating significant improvements in translation
+accuracy and fluency.
 
 Looking ahead, the project aims to further enhance the model by expanding the dataset to include a broader variety
 of sentence structures and conducting more comprehensive evaluations. This ongoing work holds the potential to

src/collections/news/2024-05-10-Exploring-RAG.md

@@ -5,11 +5,12 @@ slug: exploring-rag
 excerpt: Improving communication with AI for Blissymbolics users through advanced retrieval techniques.
 ---
 
-To improve communication for Blissymbolics users, ongoing work is exploring the integration of Retrieval-Augmented
-Generation (RAG) into the [Baby Bliss Bot project](https://floeproject.org/baby-bliss-bot-project/). RAG is an
-advanced AI technique designed to enhance the accuracy of generative models by incorporating factual knowledge from
-external sources. This approach helps to overcome limitations in traditional language models, which may sometimes
-generate responses that are inaccurate or inconsistent due to relying solely on their training data.
+To improve communication for Blissymbolics users, ongoing work is exploring the integration of
+[Retrieval-Augmented Generation (RAG)](https://en.wikipedia.org/wiki/Prompt_engineering#Retrieval-augmented_generation)
+into the [Baby Bliss Bot project](https://floeproject.org/baby-bliss-bot-project/). RAG is an advanced AI technique
+designed to enhance the accuracy of generative models by incorporating factual knowledge from external sources.
+This approach helps to overcome limitations in traditional language models, which may sometimes generate responses
+that are inaccurate or inconsistent due to relying solely on their training data.
 
 During a collaborative session with an Augmentative and Alternative Communication (AAC) user, the potential of RAG
 became evident. For instance, when the user expressed interest in inviting "Roy nephew" to a birthday party,