Chain of Thought (CoT) - Revision history

BPeat at 09:47, 30 January 2025

2025-01-30T09:47:09Z

BPeat at 09:42, 30 January 2025

2025-01-30T09:42:58Z

BPeat at 13:29, 3 March 2024

2024-03-03T13:29:16Z

BPeat: /* Chain of Thought (CoT) meets Instruction Fine-Tuning */

2023-09-27T11:57:06Z

‎Chain of Thought (CoT) meets Instruction Fine-Tuning

BPeat: /* Chain of Thought (CoT) meets Instruction Fine-Tuning */

2023-09-27T11:54:10Z

‎Chain of Thought (CoT) meets Instruction Fine-Tuning

BPeat at 00:25, 9 September 2023

2023-09-09T00:25:11Z

BPeat at 01:19, 17 August 2023

2023-08-17T01:19:45Z

BPeat at 01:50, 8 July 2023

2023-07-08T01:50:08Z

BPeat at 11:16, 31 May 2023

2023-05-31T11:16:47Z

BPeat at 11:16, 27 May 2023

2023-05-27T11:16:12Z

@@ Line 20: / Line 20: @@
 [https://www.bing.com/news/search?q=AI+Chain+Tree+Thought+Emergence&qft=interval%3d%228%22 ...Bing News]
-* [[Mixture-of-Experts (MoE)]] ... [[Chain of Thought (CoT)]] ... [[In-Context Learning (ICL)]]
+* [[Mixture-of-Experts (MoE)]] ... [[Mistral]] ... [[Chain of Thought (CoT)]]
 * [[Artificial General Intelligence (AGI) to Singularity]] ... [[Inside Out - Curious Optimistic Reasoning| Curious Reasoning]] ... [[Emergence]] ... [[Moonshots]] ... [[Explainable / Interpretable AI|Explainable AI]] ...  [[Algorithm Administration#Automated Learning|Automated Learning]]
 * [[Math for Intelligence#Mathematical Reasoning|Mathematical Reasoning]]

@@ Line 20: / Line 20: @@
 [https://www.bing.com/news/search?q=AI+Chain+Tree+Thought+Emergence&qft=interval%3d%228%22 ...Bing News]
 * [[Artificial General Intelligence (AGI) to Singularity]] ... [[Inside Out - Curious Optimistic Reasoning| Curious Reasoning]] ... [[Emergence]] ... [[Moonshots]] ... [[Explainable / Interpretable AI|Explainable AI]] ...  [[Algorithm Administration#Automated Learning|Automated Learning]]
 * [[Math for Intelligence#Mathematical Reasoning|Mathematical Reasoning]]

@@ Line 40: / Line 40: @@
 * [https://the-decoder.com/system-2-inspired-method-enhances-gpt-4s-logic-capability GPT-4's logic capabilities can be enhanced with a "Tree of Thoughts" | Maximilian Schreiner - The Decoder]
-"Tree of Thoughts" is a new framework for inferencing language models like [[GPT-4]], inspired by prompt engineering methods like Chain of Thought. It is a novel approach aimed at improving the problem-solving capabilities of auto-regressive [[Large Language Model (LLM)]]s by allowing them to explore multiple reasoning paths over thoughts. To implement ToT as a software system, an [[Large Language Model (LLM)|LLM]] is augmented with additional modules including a prompter agent, a checker module, a memory module, and a ToT controller. These modules engage in a multi-round conversation with the [[Large Language Model (LLM)|LLM]] to solve a given problem. The memory module records the conversation and state history of the problem-solving process, which allows the system to backtrack to previous steps of the thought-process and explore other directions from there.
+"Tree of Thoughts" is a new framework for inferencing language models like [[GPT-4]], inspired by prompt engineering methods like Chain of Thought. It is a novel approach aimed at improving the problem-solving capabilities of auto-regressive [[Large Language Model (LLM)]]s by allowing them to explore multiple reasoning paths over thoughts. To implement ToT as a software system, an [[Large Language Model (LLM)|LLM]] is augmented with additional modules including a prompter agent, a checker module, a [[memory]] module, and a ToT controller. These modules engage in a multi-round conversation with the [[Large Language Model (LLM)|LLM]] to solve a given problem. The [[memory]] module records the conversation and state history of the problem-solving process, which allows the system to backtrack to previous steps of the thought-process and explore other directions from there.
 <youtube>PFK5g_kxhVM</youtube>

@@ Line 82: / Line 82: @@
 # Therefore, the average of 10, 20, and 30 is 20.
-Once we have generated a set of CoT demonstrations, we can [[Fine-tuning#Implementing Chain of Thought (CoT) |fine-tune]] the [[Large Language Model (LLM)|LLM]] on them using any standard [[Fine-tuning#Implementing Chain of Thought (CoT) |fine-tuning]] technique. For example, we could use supervised learning to train the [[Large Language Model (LLM)|LLM]] to predict the next step in a CoT demonstration sequence.
+Once we have generated a set of CoT demonstrations, we can [[Fine-tuning#Implementing Chain of Thought (CoT) |fine-tune]] the [[Large Language Model (LLM)|LLM]] on them using any standard [[Fine-tuning#Implementing Chain of Thought (CoT) |fine-tuning]] technique. For example, we could use [[Supervised|Supervised Learning]] to train the [[Large Language Model (LLM)|LLM]] to predict the next step in a CoT demonstration sequence.
 After [[Fine-tuning#Implementing Chain of Thought (CoT) |fine-tuning]], the [[Large Language Model (LLM)|LLM]] should be able to solve the average problem step-by-step, even if it has never seen the problem before. This is because the CoT [[Fine-tuning#Implementing Chain of Thought (CoT) |fine-tuning]] has taught the [[Large Language Model (LLM)|LLM]] to solve the problem in a generalizable way.

← Older revision		Revision as of 11:54, 27 September 2023
Line 50:		Line 50:

	= <span id="Chain of Thought (CoT) meets Instruction Fine-Tuning"></span>Chain of Thought (CoT) meets Instruction [[Fine-Tuning]] =		= <span id="Chain of Thought (CoT) meets Instruction Fine-Tuning"></span>Chain of Thought (CoT) meets Instruction [[Fine-Tuning]] =
		+	* [[Few Shot Learning#Zero-Shot Prompting\| Zero-Shot Prompting]]
		+	* [[Fine-tuning#Implementing Chain of Thought (CoT) \|Implementing Chain of Thought (CoT) to Fine-tune models]]
		+
		+	Chain of Thought (CoT) meets Instruction Fine-Tuning is a new approach to [[Fine-tuning#Implementing Chain of Thought (CoT) \|fine-tuning]] [[Large Language Model (LLM)]] that combines the benefits of CoT prompting and instruction [[Fine-tuning#Implementing Chain of Thought (CoT) \|fine-tuning]].
		+
		+	CoT prompting is a technique for enabling [[Large Language Model (LLM)\|LLM]]s to perform multi-step reasoning by prompting them to generate a step-by-step explanation of their reasoning process. Instruction [[Fine-tuning#Implementing Chain of Thought (CoT) \|fine-tuning]] is a technique for [[Fine-tuning#Implementing Chain of Thought (CoT) \|fine-tuning]] [[Large Language Model (LLM)\|LLM]]s to perform specific tasks by providing them with explicit instructions.
		+
		+	The CoT meets Instruction [[Fine-tuning#Implementing Chain of Thought (CoT) \|fine-tuning]] approach involves [[Fine-tuning#Implementing Chain of Thought (CoT) \|fine-tuning]] an [[Large Language Model (LLM)\|LLM]] on a dataset of CoT demonstrations that have been generated using instruction [[Fine-tuning#Implementing Chain of Thought (CoT) \|fine-tuning]]. This approach has several advantages over traditional [[Fine-tuning#Implementing Chain of Thought (CoT) \|fine-tuning]] techniques:
		+
		+	* Improved performance on reasoning tasks. CoT prompting has been shown to improve LLM performance on reasoning tasks by a large margin. This is because CoT prompting helps the model to understand the reasoning process and to generate more accurate and contextually relevant outputs.
		+	* Reduced data requirements. [[Fine-tuning#Implementing Chain of Thought (CoT) \|Fine-tuning]] a model with CoT demonstrations requires much less data than [[Fine-tuning#Implementing Chain of Thought (CoT) \|fine-tuning]] with task-specific examples. This is because the CoT demonstrations provide the model with the necessary information to solve the task, even if the model has never seen the task before.
		+	* Improved generalization ability. CoT [[Fine-tuning#Implementing Chain of Thought (CoT) \|fine-tuned]] models have been shown to generalize better to new tasks and datasets than task-specific [[Fine-tuning#Implementing Chain of Thought (CoT) \|fine-tuned]] models. This is because CoT [[Fine-tuning#Implementing Chain of Thought (CoT) \|fine-tuned]] models learn to solve tasks in a more generalizable way, rather than learning to solve specific examples.
		+
		+
		+	Overall, the CoT meets Instruction Fine-Tuning approach is a promising new approach to [[Fine-tuning#Implementing Chain of Thought (CoT) \|fine-tuning]] [[Large Language Model (LLM)\|LLM]]s that can improve their performance on reasoning tasks, reduce data requirements, and improve generalization ability.
		+
		+	Here is an example of how the CoT meets Instruction [[Fine-tuning#Implementing Chain of Thought (CoT) \|fine-tuning]] approach could be used to [[Fine-tuning#Implementing Chain of Thought (CoT) \|fine-tune]] an [[Large Language Model (LLM)\|LLM]] to solve the following math problem:
		+
		+	Problem: What is the average of 10, 20, and 30?
		+
		+	First, we would use instruction [[Fine-tuning#Implementing Chain of Thought (CoT) \|fine-tuning]] to generate a set of CoT demonstrations for the problem. For example, we could prompt the [[Large Language Model (LLM)\|LLM]] with the following instruction:
		+
		+	Instruction: Explain step-by-step how to calculate the average of 10, 20, and 30.
		+
		+	The [[Large Language Model (LLM)\|LLM]] would then generate a step-by-step explanation of how to solve the problem, similar to the following:
		+
		+	CoT demonstration:
		+
		+	# Start by adding all three numbers together: 10 + 20 + 30 = 60.
		+	# Then, divide the sum by the number of numbers: 60 / 3 = 20.
		+	# Therefore, the average of 10, 20, and 30 is 20.
		+
		+	Once we have generated a set of CoT demonstrations, we can [[Fine-tuning#Implementing Chain of Thought (CoT) \|fine-tune]] the [[Large Language Model (LLM)\|LLM]] on them using any standard [[Fine-tuning#Implementing Chain of Thought (CoT) \|fine-tuning]] technique. For example, we could use supervised learning to train the [[Large Language Model (LLM)\|LLM]] to predict the next step in a CoT demonstration sequence.
		+
		+	After [[Fine-tuning#Implementing Chain of Thought (CoT) \|fine-tuning]], the [[Large Language Model (LLM)\|LLM]] should be able to solve the average problem step-by-step, even if it has never seen the problem before. This is because the CoT [[Fine-tuning#Implementing Chain of Thought (CoT) \|fine-tuning]] has taught the [[Large Language Model (LLM)\|LLM]] to solve the problem in a generalizable way.
		+
		+	The CoT meets Instruction [[Fine-tuning#Implementing Chain of Thought (CoT) \|fine-tuning]] approach is a powerful and versatile technique for [[Fine-tuning#Implementing Chain of Thought (CoT) \|fine-tuning]] [[Large Language Model (LLM)\|LLM]]s to perform a wide range of tasks, including reasoning tasks. It is especially useful for tasks where it is difficult or expensive to obtain labeled training data, or for tasks where the [[Large Language Model (LLM)\|LLM]] needs to be able to generalize to new tasks and datasets.

	<youtube>-AZL31jop9Y</youtube>		<youtube>-AZL31jop9Y</youtube>

@@ Line 49: / Line 49: @@
 <youtube>G9dRK9TNAeg</youtube>
-= <span id="Chain of Thought (CoT) meets Instruction Fine-Tuning"></span>Chain of Thought (CoT) meets Instruction Fine-Tuning =
+= <span id="Chain of Thought (CoT) meets Instruction Fine-Tuning"></span>Chain of Thought (CoT) meets Instruction [[Fine-Tuning]] =
 <youtube>-AZL31jop9Y</youtube>

@@ Line 2: / Line 2: @@
 |title=PRIMO.ai
 |titlemode=append
-|keywords=artificial, intelligence, machine, learning, models, algorithms, data, singularity, moonshot, TensorFlow, Google, Nvidia, Microsoft, Azure, Amazon, AWS, Meta, Facebook
+|keywords=ChatGPT, artificial, intelligence, machine, learning, GPT-4, GPT-5, NLP, NLG, NLC, NLU, models, data, singularity, moonshot, Sentience, AGI, Emergence, Moonshot, Explainable, TensorFlow, Google, Nvidia, Microsoft, Azure, Amazon, AWS, Hugging Face, OpenAI, Tensorflow, OpenAI, Google, Nvidia, Microsoft, Azure, Amazon, AWS, Meta, LLM, metaverse, assistants, agents, digital twin, IoT, Transhumanism, Immersive Reality, Generative AI, Conversational AI, Perplexity, Bing, You, Bard, Ernie, prompt Engineering LangChain, Video/Image, Vision, End-to-End Speech, Synthesize Speech, Speech Recognition, Stanford, MIT |description=Helpful resources for your journey with artificial intelligence; videos, articles, techniques, courses, profiles, and tools
 |description=Helpful resources for your journey with artificial intelligence; videos, articles, techniques, courses, profiles, and tools
 }}
 [https://www.youtube.com/results?search_query=AI+Chain+Tree+Thought+Emergence YouTube]

← Older revision		Revision as of 11:16, 31 May 2023
Line 39:		Line 39:
	<youtube>bjnTy2TdmYw</youtube>		<youtube>bjnTy2TdmYw</youtube>
	<youtube>G9dRK9TNAeg</youtube>		<youtube>G9dRK9TNAeg</youtube>
		+
		+	= <span id="Chain of Thought (CoT) meets Instruction Fine-Tuning"></span>Chain of Thought (CoT) meets Instruction Fine-Tuning =
		+
		+	<youtube>-AZL31jop9Y</youtube>

@@ Line 13: / Line 13: @@
 * [[Singularity]] ... [[Artificial Consciousness / Sentience|Sentience]] ... [[Artificial General Intelligence (AGI)| AGI]] ... [[Inside Out - Curious Optimistic Reasoning| Curious Reasoning]] ... [[Emergence]] ... [[Moonshots]] ... [[Explainable / Interpretable AI|Explainable AI]] ...  [[Algorithm Administration#Automated Learning|Automated Learning]]
 * [[In-Context Learning (ICL)]]
 AI can generate text that follows a logical and coherent sequence of ideas, building on previous statements to form a chain of thought. Chain of thought (CoT) is a method that breaks a problem down into a series of intermediate reasoning steps. It has significantly improved the ability of [[Large Language Model (LLM)]]s to perform complex reasoning. It is the current state-of-the-art in teaching LLMs how to take action. An example of CoT prompting can be seen in solving a simple word problem. Without CoT prompting, GPT-3 (davinci-003) fails to solve the problem. However, with CoT prompting, GPT-3 (davinci-003) successfully solves the same problem by breaking it down into intermediate reasoning steps.