add llms unsupervised learners review

Author: ajanibekcode

_config.yml

@@ -1,6 +1,5 @@
 collections:
   notes:
     output: true
-    sort_by: date
-    order: ascending
-
+  reviews:
+    output: true

_layouts/review.html

@@ -0,0 +1,6 @@
+---
+layout: default
+---
+<div id="review">
+{{ content }}
+</div>

_notes/2025-01-06-nlp-lecture1.md _notes/nlp-l1.md_notes/2025-01-06-nlp-lecture1.md renamed to _notes/nlp-l1.md

@@ -1,6 +1,6 @@
 ---
-layout: post
-title: paper review Chain-of-Thought in LLMs
+layout: review
+title: Chain-of-Thought in LLMs
 ---
 
 paper review Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
@@ -16,7 +16,9 @@ The focus is on arithmetic, common sense, and symbolic reasoning tasks. Specific
 
 <h2>"Solution"</h2>
 We have seen a range of benefits when increasing the size of the Language Model, but unlocking reasoning ability can help us achieve high performance on tasks such as arithmetic, commonsense, and symbolic reasoning. The proposed method is motivated by two ideas: <br>
+
 1. Generating natural language rationales that lead to the final answer (prior work involves generating intermediate steps from scratch or finetuning a pretrained model)
+
 2. in-context few-shot learning via prompting where one can "prompt" a model with a few input-output exemplars demonstrating the task (has been successful with question-answering tasks)
 
 What is few-shot learning?

_notes/2025-01-08-text-normalization.md _notes/nlp-l2.md_notes/2025-01-08-text-normalization.md renamed to _notes/nlp-l2.md

@@ -0,0 +1,16 @@
+---
+layout: review
+title: Language Models are Unsupervised Multitask learners
+---
+
+paper review Language Models are Unsupervised Multitask learners
+================
+
+<p class="meta">13 Jan 2025</p>
+
+<h1>Overview</h1>
+This is a blog post created by OpenAI in 2019, which examines the potential for artificial intelligence, where
+they showcased how training large models on diverse datasets can lead to versatile systems capable of performing a
+wide range of tasks with minimal additional training. The blog post can be found [here](https://cdn.openai.com/better-language-models/language_models_are_unsupervised_multitask_learners.pdf).
+
+

_posts/2025-01-10-paper-review-cot.md _reviews/CoT-review.md_posts/2025-01-10-paper-review-cot.md renamed to _reviews/CoT-review.md

@@ -11,10 +11,17 @@ <h1>blog</h1>
     {% endfor %}
   </ul>
 
+  <h1>paper reviews</h1>
+  <ul class="posts">
+    {% for review in site.reviews %}
+      <li> &raquo; <a href="{{ review.url }}">{{ review.title }}</a></li>
+    {% endfor %}
+  </ul>
+  
   <h1>"lecture" notes</h1>
   <ul class="posts">
     {% for note in site.notes %}
-    <li><span>{{ note.date | date_to_string }}</span> &raquo; <a href="{{ note.url }}">{{ note.title }}</a></li>
+    <li>&raquo; <a href="{{ note.url }}">{{ note.title }}</a></li>
     {% endfor %}
   </ul>