Learn Before
Formula for Final Answer Synthesis
The concluding stage of the problem decomposition framework involves generating the definitive answer to the initial input problem, . This final answer, , is determined by a synthesis function that utilizes the original problem alongside the entire history of intermediate steps, including all sub-problems and their corresponding answers, represented as . The process is mathematically expressed as: .
0
1
Tags
Ch.3 Prompting - Foundations of Large Language Models
Foundations of Large Language Models
Foundations of Large Language Models Course
Computing Sciences
Related
Recursive Formula for Sub-Problem Solving
Formula for Final Answer Synthesis
A system is designed to solve a complex problem by breaking it into a sequence of smaller parts. The process for solving the i-th part is defined by the formula: where is the answer to the i-th part (), is the solving function, is the original problem, and represents all preceding parts and their answers. If the system fails to correctly solve the second part (), which of the following scenarios represents a failure caused by an incorrect application of this specific formula?
Analyzing Components of a Problem-Solving Formula
A system is tasked with planning a vacation by breaking the task into sequential parts. Match each component of the problem-solving formula, , to its corresponding element in the vacation planning scenario.
Learn After
A system is designed to solve a complex problem (
p_0) by first breaking it down into a series ofnsub-problems (p_1top_n) and finding their respective answers (a_1toa_n). The final answer to the original problem (a_0) is then generated in a concluding synthesis step, represented by the functiona_0 = S_0(p_0, {p_<=n, a_<=n}). Based on this formula, which of the following sets of information is required by the synthesis functionS_0to produce the final answera_0?Diagnosing a Faulty Summarization System
Consider a system that solves a complex problem (
p_0) by breaking it down into a series of sub-problems and their answers ({p_≤n, a_≤n}). According to the final answer synthesis formula,a_0 = S_0(p_0, {p_≤n, a_≤n}), the synthesis functionS_0could still produce a correct and complete final answer (a_0) if it were only provided with the set of sub-problem answers (a_≤n).